From d41910061f7307d040304ef4c802ceb4194a5162 Mon Sep 17 00:00:00 2001 From: Thomas Fillon Date: Mon, 11 Aug 2014 23:42:01 +0200 Subject: [PATCH] DLFM 2014: english correction --- Conferences/2014_DLFM/dlfm2014_Telemeta.pdf | Bin 3594732 -> 3594136 bytes Conferences/2014_DLFM/dlfm2014_Telemeta.tex | 189 ++++++++++---------- 2 files changed, 94 insertions(+), 95 deletions(-) diff --git a/Conferences/2014_DLFM/dlfm2014_Telemeta.pdf b/Conferences/2014_DLFM/dlfm2014_Telemeta.pdf index d043dba0f57878271f116f371eec44250638831b..7232be53ddde279d68bc3aa51bcb55257c2b1a8a 100644 GIT binary patch delta 62929 zcmV)BK*PW6v)cifuiF8TBLgrwF_(c+G$?=VI?Hz3y0!Csg=VGJs(1+S-BBFJ>2a>@ zq?WSUwVDoSiMDwniaMkeALr}uegOn2I!)Tao!o&;64(It`vnKHhgmTD>#N}3Jm0Kd z{qinWGnM*zsI%GXex_95CrLa@+vMvwBU*?N2ioCPke4&5S z%xm`LZqYTzcCG_&UGm^ldB-pJZBdtV6?;!O-Lzk1n7`)#TD_kwLgfc(I$NmR&vncP zmWRh(QzO)|tJcl7d8pRKmKPVtP1VRNpE_58_gH-`J07izU4cUn^9B3`q<04a-)-nf z6$O51k4*T;;&^E4;!v#(ESv|3KX-pshl_rB7pAk!&yyq~HYI_taz!I4Z}2$?LWYbr zu+h(Bcmn7`C)!Vem5Vs{RjP%hk8qkG{I1$=o881IY3v6{!UMeQ*`5HXCx|;75XF8J z6E%?*WUkJx|19i{eI06;4g4rkJp1Y6^*JB?X;*!nXMyLlR6ow60J*vV$}4}tE=Xvj z)Q{t2w9WM`gPh;oF8M80c`(fcc+XQ=&&v$8hL|1eWO9LtWdo+W>N5DxirQ|u09+X{(S1wY>+>G$&xG?GY~@NNei&z@(WzmWfh!jy90~91>i)PZr}P~T3-S@VD5=Hu&98qq3*x?xy|V49 z12T=b8%cqX%KUgt_S8GSpv>5O|7XT^b9CG0JaCS-YF|swx^#{Bf=z2A|ztE|(u}SUP}_RwLi5 z#4332qa67j<9D(C{b*wMRdZzH-7Z$NaoKfM$pnPi>9+rkIlzB@DjevOo(cm<3SGC$n&NU zKgD(t!T}>i3RRG323{ezK76|R@M(D=cwW4{_y@n{oG`@7*ZBy8AMq`aE(U5q+O}Lb z-(FY3#)EY)prwC)23s#64T}Of;sW#?RF6?Pt10ywQo1c@IV2KUIMntxZHdk)05zHl z7m2U3|93<`WD@8i)Ch_44%k})Z(l#uMYZisA&LDISnN2Iw*}hs*J>k(HpegE!lq=B zQYJj)^^e5u#qubd|H4Q>N*#tn;bJCmb$PwK6cPDZJd%IVhqf!`6t`3m{EWv4{fJ}) zduGXqW?#sCy{+o5vR)UG{{%>lLSHADENSX$SN(3B;5$SG+&jCuF{iLZ89Aw|;?3

BGD^H(;PjiO-IT@g5B_o#JRWHTtPq~90qyR@Oj%p2JKUNXdZ?Nm zyO-QgPqu#xqaZd83VnXDEl~nLqq&$8j<&tGY<;~g?nHq^EGLyP>=zA1jl~*%rOS{081Z*a}3iyldIF!xFQ)MPR0@1XmtGnsnALaF~LhMC{SvdB>|-o)fY{uss*=?Ia(1WSF3~M3TC4 zil#!~&^LxmJ90AAG+I9O!>$(M*n!0WIJ-RFHo_D{pp>mhRVu;(?J^F%-;X>E+f=f0 z*B_1WcUBvYq9<~uQ6qp89!VISplmNcl0$z%>=1krwJ1%zdlDzV6C@#3wqGl&zzUs7 zj4nj%F7NHGhoT)EZboTrj)nzaaluAbIScl4P4wL%>sSG&4Qeev$ZBCFIJE<@4PGO( zat#wDPt-yx1o6sfbH{*>^iAlXzhHP44pEJWftsdC?EM|r!p3}nu)UIfK=bKY8!La< z5R)c^@G~}Ow5B_-CPZm8WN643J~(VF%eLdGQ`%tpZ;cCUHu)T=5D#pTL|NWhgPAlF z4KsoKQdGXgJ2d=cM`?Vp@g{qoEcnq`zfu@s8!b+UCYTKU+|Ux*!WfJp5*Iwt`yGGm zWp9FH<*s=~G<|}&+F($OxC=@ehJk;JAbw7`mWc%eg~dTpSbQ~xKxdKn`X=%TL4chx zdgsgu)02>g8aqML%D|yZiK&g8Gn3+|XA+*>GYMuWxrXPRaHB({&|x(@ey-+O%u+6~ zKkXd~_76JM!Y5kq_Noq;Ee31jox*+ArNs-3Q(6s459r6CsKc_gBxR1 zg@gs-X`EQ*$H^)4Io{FS3G3td#cTt)gE1}7y`}L3B=GOqkCGp8EV22Cj`Jzb(zp=K z(MV^OZk&6Z_aQ3@Ogt!eJZreJ=ePtsdnPT|@H82)Y+{^{nX++af|k9p`ZlI?lZQBmsXuPABvOyv<7{GU=P244f>2EdrC((R*#Pu--M0LPa^S zn%GHj16l*`mQR4meo;DUdtiB5ZuO1$lO z(4+u*qz(OWW!ymB^>EG^R&tedPRE|g@T78m2`;KjL;qrmjE&=d3Qi~zmD{2w%##V1 zasSB$S1{Ya9FXdVF!e-`HB_v*j}>b$+s>A+{>B%&gDXU{ zSCUX0VsukEs9z&L!5TXB^DI6Mzai9rL!*<5Pv-WbDD7cOR!o2Mop5+>C1MLhazU`p zx}Xg8mL3nr(oErqgZC|a6Z$X_hbPKi@BX*}&6GHovp24-H;y-ikQoK}q_>NLjOx6H z(n*z`8UAw+2RKIof!cut{$%rQ_`v3ycV8Ox2w_}%MvG7}F@|c_0-c-Q^A8C1cHyL~ z2&(;wd*T2aYFdAXgFE7pZ0W}A1=;GfHwLSqEss~u{MPyu{)QAh*^ZqHHf$g$bCYIp zXRae(sSL1{)_uzSMOIwGk1h}>sR)5f$eRQma)6y5LapS__qGa*wJkxpS*7ucAR6%T zYL0o|kpijWdj-IrjoK8rtsI*q-bYdau9EEFJ4Dfl`}}`mvV+ujk_L>|N$3ZF5~PW~ z&QC%_m_*)tDR?9uSem4~+H=W_Fx$c@WRGqnH)y76A{9s`tJhMj7G^ibW3#^-G(>3E zz_aR$AQ^#7MvOsCtcEq8l!IJn)cqkeWis8p4YzKDu56jFP_2Y&8}g{=*Wj51{L!4w zr{Zz!m3V)ktY1Q3vd)dfgHeSS1)=xrN|L@S{vPLUQJ_s$$w?}w*DZ$5C@_&sBrJlU zjn+0pv-@s2X4yU4UOva~W$0Fm+R7$=>WYesTPEA5bdML1GBYY9q-kkU#iCVSmCtdAn@i6WVNTfHN^S3W+83 zNm)93<&Kw{PKr1br{ZRn;&pDIcJz@o`P)lA_@w<3sh9MJlcd4TQ-J>f;472NDIb6T zi!lhi%bV3L-e`Sz`{{gj$^U9G$~8Fa!K?8+mQ-KvDIzF8Y`>>{<^faQwht(yB&C4k zmWx{%rh&5WtkS405KW{kB%>kBB5 zO34uiYWm8AmY^TUXJ=cLcq0%6!}@<2Ax>JUr5oz_{Hl{Q3J;ZAmMQz}6@j;_OX~=1 z{h_~elVUOgW5@suVPjmW=>ATmu(xYFnM~BIhY}(lvKm^g!!c6lu zfGNkOnr8=@Lmq`;ILz>yZnEz`)t4V%i8=zI<}lP*E2ldcnp>OA$(_L9@=bCsLV{15 z3cSW|RHkp>UOjN5jL6eWUgv)lK%_d~3|2!@(c=$tS+}1qtT?*0594;M{e9i{XZ*2@*fAbafsKD;4DWs3$(=E#if5-Rz#e>Yo3p& zD{$iH@E;%scEp|4=sVY9Qw{0w6(P^q+QBma)+E*3-U$OeWUcl-qAdGcacYi8krYyY zlAzoGNU7n@bi81)>#L2mRk_6r`O&x1)YCa%-ljPh%chpjPcCL@aM;VGAzNXjP+{QV z2U$axUojE^T5V<_1`2%P=&>Am|PyxHA^H~;=Dl;6MIKKt%X65W(RmFMZr?e->1f}$#KilPjn zxVpLB-2Ad=Ps?a=TE>g>GA z#RV00)CvCDai^BYuc)@^HrS?L@?h<~j7F=#A-YExvI=R;dm6pLsule8Sa!iS*tOwy z4>a2yHTmuK@38Sz6b5M?@$T+q*d7 zJ_L7_FK(A*zGx0O^pO@=%TYjGE8RB)KNzju$?6q6P&v@D>G9VvN@)^Pv1iEZ8akYM z+w`>Gvy4>}Xx&}#Cc5T#Il#7S2;>GWV1He4o)EOQ{Ae+2*kegdFg;%cSvTCu2OXg- z;TDbg9ET`j+A)ZmuIC|ZXQ(6A(%H4wkZH{r!n^DRtZ}%|A_QQR8UVs$t;BN1(I2FK zF91iBiZZwT12&`M{N&a#`WSR;I<91Z!(+n8(Cy@OaOR`*ytw4o&uvSfo`m@Xfq$Wp zu|;vMCJbGrY05a+a1%k4t{?c_;Ko4z;Pj73%{h0--8lyrw?m;lozl`CjJj81=7x+S zO+y>JvM)TALu-(Hl@tZQRA(m0LQ}W8)C2Onuu@g@554tC{70 zz>zkY`sf){Y>&qi!nA2p9G!ynlkHRwWSp_$}+yhL;^6cM`L}HOT ziiileIGYo4G>(Pm6v+g*ihst(hJ~<5d{eCKhlXuf(6gW(BhXkKc&Wg?mxjtPNtg0D z0Cj%-wEl2a26+0#3q9AwW$@Z;X@bl|F!S#_h*^CGK_StsxBEea2| zN@ZzWdMX;cidZJ@v6(bB%zhTpzM#|MFQXq&B5C>wpV>*JgRrBxLaxap(aUr|fpuJ^ zPcUETiR=!49a;sZWq&cnS}arQ->cc7ToDb8H1g0O1x3oRs7Mzha%QiGVo%PcL2NIu z#Zjl51PaVS7T!AJ&s!Nf^MNsP=4eE_f`xrV9+dvPApw`doVzr}=(0^73$GYlKx5_Hwb;^+& ztD0?j2(#C|;etDBY&OR3F$-Y9f)H13^c6bVu`%#*CBTWT%~cwwNG0^!H#~5s`V0=A5;2K)eNSr!pWi8n;xH)l zO2>#~kpEfc87Vg?KkCZ{bxW3Zr=d9^!%u(_)~9;c9Dn6KMFwP;kYwg9`^gKSRs0gYghTj-=9y>A^;RV=Byrez4-j}#Ec+**0 z?xn+d%S{|+p*?{5L1kO~6A@uOaB~s$+T-lZDjbaTVQzt@IIX$#eez4zY_%?%8rU2P1h>9;1bJd@z5?+y`lG-d-R5^RL?!g&Y?|DGHMWm9!rL^ zog90vN$cpJU*+XZPpc8(1+ujAI9^y1#jA4LNt521_oiKImeO-UeQ*eU$7Q_By)+Ph z(vOs%LAVsQIjTNf&#DT(gJ2b>iNOvorlPf32!Ap}wM>QoV1rMz{YQS$Iw9eaAq`E8 zM(R%2%uUt&VHe?m6^9_hc6XAsY-e|8R20W$5R0OVw2*{8EmUwox&Zces(3?233NJR zyfERWH(5{>MUI3UB~cKS(G9YwN?GufzSAJeqMMb|3i)p)I#XQHc>q|qUDIfRg@EkB z_#AED_1-%YY3r$-kMPg``i-XYus77$I>eQfzFASShDKVy zs$w$lS$Wc*GGRDnJie{nfsW>{v-9)DC#$h*KyQ>sOiIW*j)6(ddtTR-LLH27<9 zzhQT$%OvsLN1-ID>?vcl>T@+i1vU1K913|7OY2FDh3y=D14(T$7)%doxF{o(1#IWG zL2y^|D~;nBi~|6CR;}|Qu*Q~Ggai#DUgI=Q7QN$5pt`46DUEAbsVsvU)N`v>$A3_4 z@l!Ry!Qr^d64N+$rHSiXPiZwGQfu{BHj3Y53qWsU3ydVR0!$rDvDLuYl;~|CAM%u5 zuF%0`6Jq=fYMjxJa)WNyZ~;-Y4%Cool`meaC{xSMn02X{>&W=ozWY!eT^NC3D*&R;RHS>o* zi&&)Z_sVB<8Nw{qOsq5l8!7745`-Y?_5(O@1VvbYqAaMGYB-UM8an*ufu~jbG0r`M zlf8Nwo0|=pP6Ww+y|xSMKuicXO@whv(c$pDJcfxbAVxB6LLk%=~Eq1CA3t$}eHO1EOr5p#k2G@^gSRu3{ zms~C$OF;;GN~F)_g%JbL@Qs}P&K)I|2t{I|JX3-`S`Ch!fK&TT-G*C!oO9Y@Pu6L_ z$2m;w8gz-m(IuiJNaBpF8GqoqyZ``rG4A8)A5{=m5mZp)pLBZ}@ZcZd`SuUh%0e@6O0jWC;f%pJN zOmfii^D--}e&#w;)7Zv(52NtEMau2DJ34FXzy4(Y#KC-+^*yIx=9^^=uAIn1k}Rz# zVffxVLQSZh!a8Y%wy=~K&n=GAs{BF$Qn21n1E52ir*N(7Sv^s3hojpX5swPtXa`+4 z#ag0}4dcQrqnFk*{(pLEuDWd$8MGWYF6k%^Yi>KMAx#}!?uop7Y@iM{9VAfoKrN3qQB(!!Lke#WHy@t;m!#Ra_E84*4m5~UIPNO- zj5j!Z-+1$Ex4kTU5a#_o{Cy{XgDCa~lRUt`mSRLdARL(*ZGS7CEf0WoT{qe$OLAqh zRb5vx=p<`P;|-Uva}E_jH#!>+%JYrgS0`yRo3@v?cj})WQk&n`6hU&D40}#iATEIoKXEzmG$ms8rLhO9>H71DZA=*ze%FV(@4NGv z!fQZ#1TZwad;Z-4uOCRCNYEl2_UFsMFWaW!KKB?6$bZnor|pJ)Esf015_DxY?y}^( zPG@x4fNg%PHw|a<;M%Zc)AmEx-YvoOyU-f2aI3X2?XO`N4}PEZ`FL!)b&LDN_j&`m zDcT4_y_4bDw~C$vS|<0RWln*1*VH(0S_-gZ;IpvecHM9(k44oD*KRL(a9Fg4BOeGc z@Qr^fZGVU+dl*w94ZhsN>H_9)K#3tq2^`{lk0<*mLqOl|q3y}{jzgVi;0gA@lSyqo>}U>Lp$W_K5f3Ej`++!fMnv;hkq0Ch|oEtS4v&j<86D_;jZeA5RV4Y zLEl~ac74V`3Ly3)mpuU4^n1!8fwy17EPVBL`;OmYFqEJ~dAiF6q7np2X(apG^J#NF zJ?C!!_zz}I+Q!}UOF7Z5-R%dR<>Ns|+x5EFw`-uN@c@PBu5@kUc z27gbouMxwK%PdHfn;hg`kvfhCbz!BF1+a73OGX6Q6Bi%rdTH_H%mqnxLa>kEY%@2Zjs~LWJgQFt(ez(!g}rh@B}tylM)SL# z>%K?dPPKUD#c*+dlxsx6R+JkmRdhIcjw+&hQu=h>hVDHy4-=&lp|k`zvC zVmzYa?0r!J6%3Ib-VoJU6<9~t~6VxNz4Vl)yo zFT2w$G7jP_vM%sU2*b(JYaJ0n{@W~?iLPjJXvU_C4kXD{^nNl)#pL>zJP0s*XPr>* zt^{yaab>a~$oh4HaTJRW#eX6`I84h}#K7cRC)_>Qw^U8 zk-5kcjdT5-EI1P=QAD#B8R)rd4&GsmFEgZ3=H|!U+tjI!OC)kJxv4?HopkZ)@hNfd zAf6N8dgP$z-jNs^H)TlTuUC-(-zX>0FPKN#%g~-G409lV0CQH1s~+ zl|2^nOH!1KG|0-513s_Mp07_|E*ia;E}EAkIu|RIl_9dEiWi?w#)WrKPpIA`saSui zJ&|vKS4B@pH(bFKRV-G!)|{@dwMQnU0v|_R-EgeB7JII0$y->Cdt4W+5_NumvW;ga zp3CTC#K?-pKh-&zSAV0>Dv`7sT6Nym4se#x0n7PtODYxO|iyMJeNrx$HFHv)+Nkq5WA zR!Kb4oM+>Kk0v;LWE96hQY*hE0%fmRxJ);TL%?)SB*Es2Y&3)oWU0$@E+a zZ^&A>j@7t{v~uo-Sy_IzB#lkR_6`-kR0nr6n=X|HWmvj7Ny&_k1-3;|c>ZmVe}Nk3 zMIEc4NDC7^TYtv6HL}FQn(?5Q0qFgq+B#%wyf^7FY#T%vY>0ihz3?Nv<`%8^s(y6h z^n;sNV(v#!hN^S#PFen0vg^~zBzTa|YxYPrgRdg%v1S^LOID2rH|@aszr2HpL4Pgc z)IEswYvK%nrU1Cy7a#@4TBK7Z+{{D`f#-$uL0RU)a(@v~!dle&8drd-Pp->I`WyBe zeMwux5ggrsRke9Bzqgl}r`f8N#QKjE4FZKb949xz(;tKH%fg|)s7;1GkFBOTaKq$& zhBxf$s(>8v74jKuj6(r%HCR*Xq8p@yAB1z6^`qKNlw~v}W zh#)kx89OGKq&%&z8S|QO5ru4VNZK5}AC*baeSb@~Cuwb1*`7=q&}F2I%0w zSkJ`0M&&_}+G! zxA%JKFzaRg541uZ-?U4J}+*<$S^pg;$`;6-)j@?h5pS9vZKZkDmm z)AM7ycG$w%Xx_p=_Hs{pHhx5!qn``zrOuM0QW?N{lzS(i{76midwtTchgd7Aw4d2f zw)V+feU`SG?2OOj>!eWRrrWsFK9vfL>Cu=9eLkrg$*cgtwbxjiqPFqXeCj#j+JB=o z{-8nYf!4rvd>bykdZz^YpKOA-r?b&4PjWiXAH@8J#{UMDMZLqXnjUEPerP{8e501q z2$B$U&g;YD(5@*>11efak)ZwmLzQG@S)%9XLRbpO(AAj$q2>m%2k(4Ny%zLwy+ z$Dz#E!aNss6dNn7CtBE|k1{X3yY1NOT6&|IK>h7eG#gdn8_E9Ozgp{#vupn1%<&h) z^G@%O-Z`p1l&|Nvk5%SU=7iPH==9h@Wxna%S;;Wj^%ZH=94A*o#86rJcYlFI3S?3C z1uKYLJYcx$XIQc*Hy7;`UL^uS+IGG`i_VPhLm zLJh;q;M7!8&EUia-WND7^M99IEQ02<_M=X8;+s6~_XBd0H6gLDSseS)w=o3X2jeVv zvjxU{9>>mtL_f0j-^&8dw>NfZw*+ISx#yoA7_laVakdjb^bMx{?2cL;JP|Y;Ig;pm zGA!^82ltYE=juuLAsD)V4!_gu4p11~i7}Hz9?yiKB?$XLV85|7bAR#1D%DTC5N?Rl z)m>%Kvzgvxf`lP2IrLah8b))sN2r;-ClZE6b>7*2_Hs&SV~-*8=`dRK-gmTwoU9)X zR(j>?*5Tk}M$`AJDBldfTvhWs;o?DVOZzU9FmpeVAA{%@7=wQ&wg|mTefU0qwQ3?vdTy=4v2a+L`%$2)EUgHy zef}i4p0}Q$hct?BqAU%Pu=xKat|w6u#6XKtS_FBSUH79;+<&*$W6uivDz9?aV-N4e zlP+g$|D*S*p6O>@p3mgaM`;yQiTc~0N}%7kmbR1RvUU>Z8*}w!g#gVGKZo4r*W-{= z=Qp5RmU51L)E*A3>$KFO2sUgfi^g5Nw*Nl|7^ zUBX3YjVhAu@zt-dGNbFdq6yJ04XM1Wv#OvMrCed)&Bvok-ko3l1OEN$X{RQCQ+1ta zZL>yKIPa%l@?-~N-zRxyn>u-IzN@aAUj-j1PF+L}#T13D;Qznk0e)jh=Q)jD&XIw*y{(Pix(*#?1(~ zKzo?T%U$1mFe)K(7I{u}US>*xIj9B>Qt%!hkis5*a4Xm%HCC?_-2Z8v+M-xpuj^vf z9#CIL{J|YJJnIp8Xm9WVXn3aD{ucGU#o3E|E$Fo@Ksz=uD$i=&h+=h7i8uQ@c4k== zlxdz-rVvge%B0PSWK`aNWY&VmHzq5reL*n#8w3NFev6Vo>qfDRYy*MRl2E|z%B-vn zuv%w@vw*jLwq`O5fN_sxYd*tP5=*XaG5LK`_F}wG;^?9A@YPA!htRWb! zka3)Jpv5Dv#w~C8cEtC`HpNjq%$^=u-QIr5ax1Vd2rZ>z{*kSJVL&gZowr?dgcj|( zw5vyw(H%b=WWo*2C-Dy1IxVzaT~oAR7I^M_BEBw2>>Fh53)v^XlxZ!xVQE&wJwIQ}5 z4@mnz#Ma2~M;0D`F2pmGj!Ms=<;$A1=X{UX+~Vf_68o*=~jYv|-m#5CUkG0lt5?j%TMcnu>p*rIb$5|$Hxfec=XMI@l%L~bvV1g$)_ zkp^&zTS2bBryW>SL_fIgj;FJS#OLB;on_iIQ=y?oeb#vh&;1=?rNEbBN5IM8?>qLqtLoFUwVjW^!;+8vEF>E%;n_;_F2u z#l=b_^cqcnwwm||jJBizaayspohQxmi*4}CukXcnH|)P?H~ovT=^(Mmlw@x%_P_vH zHN172ati1W*D$n_we2^U+1%{&NV@FrvknXwo23c!+n= zWX6IVWRO%60S}tyqU|X>_N|Xb%WzAV25J?6oE0GX+Rn0j*Tb&FH|+Gtq@l?AYBXTa zh$im%akz6%*fL?i(~&25!_;zE(jKA9jabKj1Ls;wtX9>Ghg!x%3m)^Wer!zRIWva+ zf4g;ik)hs*+v4zT&xTCv;$R13!%T&aTm*+bCWF-;8w)j8Ph93f2ebCUAtM1h|92q0 z^FtUjBJUp;-P^wD^6kyum>g810qKs4%)b_E0;^K`6{lJlfQWH(QJ@BSS1*|>PnH1l=8K13zl zVrm!DsnK-MVjbcDYyFDy(}@8`?s5?v@Hz<&f(W60RtoTAp#?wcA}d{c$3QevIh^78 zTt9SClqHO+ELHoj87ak9 z3uK=u{&8p)@FtZ{OuP>;ZpjIec84I;TH`aNHC)fb6<7a8#QRYH!-)0S<2XSAr(pSn z02(-hgE$BSyB+voD-yOzku_C+RZ?@LY%*OniOK3*8ePSLB5bA#>^J&JvWpYkMoL=C zr7=J_$?30*1Y8U4)6Y1$ZpFkd1w z!7`*~3|E_Z)d3R*SKq{n5C$pRJPX!aP7n?gL5!aCbDjiNhh~n1F&B$}$-2AI7vkWH z6-1Y79}z^e(&s5XF^86lJ$Zv&E7#U zo-;<4UI3zrl1Arpl}sIfCTTnses5E6eQ9Vd`2{m8TR2SA=36i@I{~Knc;u87bt_L| zEw%>CM+}^zkv)$%o$g)#q;q^92PQl&`xC5v?7YSn4)Eyk$qjb>^hE((Xz55z{2-t! ziJG`n+F)>Rymb%L7D03@x0whBA6;F;#l&$oS*p;3c;3QUE?dZd;NgZZox}GJzLVsw zVc;pFLkO4HZNsZ+=vxww3H@ALA_3w`gt80!jF<6>&vKzktYRWnQsLq_*3yF5NM6Ke z5LZFgLKPJ)5B(j2HXONIDYce^7%rRDR#9rkQBKWNu|%o)WlM9=sb>lI&jBCRWJY=5 zqXhX7#Ch0t1lDtZaH-Q~-vR;~pJ}Mg(27@v_NU~ILjaua{6kd7w0wW-3q_W!^K~3^ zou5vc8n?bEX1f^ZYerli1Gm*Asha|maGe#ZbRW4?HD3}Ru*vcUzrN69-=qw7xKI)L z0L6{YQX@K(PbX3IGQ}i%A*p*I?dlTp@VnU>Z};_OH?F9Eyc@T}Vgq(sGBMze#NP#U zL2;7IEqh6(DNAT#ivP1vNwYmQ+tgA9*QH)+hVmJ3F!UQr#huT)D8Fre`%tm<4KCq5 zo5*phd;uWGioU=s!XFOK$JZ=I_dUjaDC=GJU}zBcv;Q!;ev&gD96(Hk$B@&tDHzk- zPp~zW?D$!KFqh9AB0}y(?gB>8rksQ;Am?$%AoO`V`VQnWt8@%t5I-N|bqbgBXL0gE zxD3%GOqm^Ym z&T)k+1usE{3!LNPq)%*t!Ql_Lw;xVYolhX2q)F_5a@fny2i7#R62bN825A|;By^7h z&7}Q~kDMwuvQ=n)Eqz0zF1F-u_hAO(Ad6E6%)UOp)bk{wFXd`V$xh$sIGj`x6-cGl94gHmb@rWN)SG%+%9b z81(x!LM?nJ?2ZyTIPt308~MMF(%oH@(=s1O*wNmQWrSykEuUoUTId|by{7>0D}Rge zUt%(}otM!xH4}sIHHYvu0f+E40*CN51BdW61c&f71&8o828Zx92Z!)A2#4@B35W1C z3WxAD3y1JE42SSF4Y%+$4#9u~H90agmw{0E;fsF@z*Em z=(n=}rwg?(c~aKZqEKm~vtqH{J^ABFqOzjMu|e>NRnmVq zRevz{U#^}!dzGt2mDEL{7gx6nW0EvA3sdKb(rR&av-oxS>Ow8Mjs5RrfBN=fmFsMo z+vo2uR@#{5RcD`f?MG;4e{Sp#H){9xdBf#x-;j-a+LNYS#9@Y_4sB5Jv9cktn0 zv3#F3Um<+aVRaS|yhqKDF#HfrumM7VT{d$@% zpI?}CdAjKi7pu%Z-L}o~)L$t3y@5*FFUP)FAG^aIJ~uiScq;8WGW$6m`Ul*cRex5d zcKLSCaNm0_r?aK%-%^ua(D*Rk2m7w@OjfcMI1^fmoHKU+3bW~ON-J%TiVo>cR)kc5 zRhlPA&9+EX8zM3&So3uId+dgf&3}16T*4hs0Lphj1Sgq-WL$q6HGz;ac&9B6v3Bb?b$PN!XQ_uG}M2#auAjx<0MPVO)@ ze@m7lKGeVxU%1c`s1O5Ff|Vr8Ev(=gtNTvoYgaIwa8=0WSkUm%*~jN6?tep6MHH_W zCaKG^0E$;-npiqoY{V4@?kK{%%5j}X5=r)5^SZIBp`NNa&GHDgD3ats{)IVv96l}51)bfgVGs|n3R=NPlz>rO2W5C~Vqc zRkduK2(&+62Uk;vtAd7ZRZxdQ0A&T76;yY=W#ikx7@dfITC-K28fzv50oQFo0+>2h z1y};|9SA1b(DEylF5mKkfwtet!hk7Ytk?GOC72loXJr(yZdDUO2GUR{ z68an`<<}Xd<*C(7xPOyWj{6IX|7WDu_SCeAd&;Y*hGSrWII6Sl@T<2+*jqFq8nIz= zI?V19QZW?_2Og2s3r!LA-RX?f0vSS+W*whfh4QOG$%A~D=XQsRULTlqu3s5b;19q#3M~UjtfJIwS zR^6|e!)?1;l#@Fc3AgV+Ys9)>W?5w^q9;{Y5yy+{LlMTBmJZTdif%Vo=k^k>02Qa@ zWs=We`^u1a^nV_luE}aKlaZe~(YjmeW2?VXt9y6s$TBzFgFkWYT2{c^YRNVktrL_% zX_mK*h#L0+XP^vIq#aBYxO(1NwMR5ELba(xoWkRdwwxvj1tV+6mX)#Ksle8&LtD@F zo3=(x?d0<$pKip0>ttCw>S=Nv83K96ink*}RO;{fXMao9__?$(c`i&0kIl&142lE) zigf6ZfG1*wja5>zYCO1eZY2isN5C#prZuS~t(`j3qPgLa+8)pONYn|uY#ZdXB3X&( z16yO9Ar8;T0z(_mz3usX-JZr;ADrz(OJB`V6@IV_NR*uu5ES0VmyJH~t{>nG!35Ch z@`0TWa(@k3iQLh`VW+NWn(Wo`0zU>O@EW#770Vw_ z0eoQJGy4d>rQ#ouo0*Ch%H<5=_BTJ&QMZJi5WO-GA_?Khh8@D)WxK?f+YvkMu2f9Ri0$&DJ@-WV~>|vN}hinGaW@7uv zXrMN`+uIhmR3IgjJ86yt=c&!sph--I;eS9SJW-H>RL}%MJR%LKBR$R1q!XgP7Tq}^ z)*bATsd;J3esdZ_8D0yc{~lZL_Lb2dl$gLUTGj4s?(tkvSZ(eYdV=~!JUK@1?AQnr z;ec{}_^FBWxE52ikr{#-VFm3}V(TRd`)C#UXk#ljvwHt1`#jJ{lShZ!la;~w4}WYB z^0$SXv%uHp0lgbHxgc>>b+#OM@P_?5l5w^|6#cLkc6TD_+#EhLWCC|}&NLRIQRz9L z6so4F8fpR=3a|L-fh`mJE;5y^8O4%G#ygM*fNcXaN#%}ajWPkt%?aHz_m7ouYz(qPk$GN1o!oY$-wR6SMtKyd?*SUxiHH&&d@;NV*4~( zYVjPpV#y(ZVUL{%p^>~*S(%iTl6g4gkd1dHQ-$rMBIf^1%YyOGtjt_@( zpdGZtb2b6(-toZ!k_asXA|_k!7$7U{e@uaSCQVeOLxrN!NsUsa`3SoTv48i=t03sX z0gWuplgy5?-aYyC_jGY%>nw96xvhG@mOE0-P1_~dUK_XsCwtfI0FR9z%O zON=(!i9G_X-SMiAY>|F%f1dG=NTTb&ZAnC5vFLd=u~{qnN|Bf>i{QwAZ7RcPI-9`p zybGq51Zs50ZU4d3N63_w(tmXMqtjOrNcr#YAL4 zl0g3Zc}rYFTAX;~9k0}{!xMfa->j6Vv=;eh^>t}Az`MOT#OwlKXn!FLnsAgNbPF=) z*{+l^iME~#>x&W;|x9TOR^KLLcx*73h=ZprNPCP=TTe6%oZ%W(R z!H2-mZsXxR8C)U$g6jp&j+=(4?C4pj$3a{_%WC0l7 zZ^)VESaI8MHI?p~BVfJTVJ2oQ-9vb0Tek(^if!Aror-PSwq5yR+fFL1*tTukww>Jk z&$*4av-+#C_nc!aodKqOn&VoNMEufGGd?A~#-FW;aD3?#(8G(=wcJ?NrHBZS22>$Nd{)W;ZGNJyZiv5SA+}d62*?()xOC*B#m&HpMWq_e+ z=-dN9=AXf{gXKmCg4v-_U`(a3k1g_c=jAEqrJRF~iHx)6!R7G@G;~@Q+A@W_c>sV8 zfoKQl;aaew9EnU=^7o4;u4t;~HAmW{iGw9W3l*LTp(J6C{B74r!O-2$r3AY{AeQG1 z(1do|A^`tw6BLBHyLH3GYkJ^A+AqxH0S^Y)7?-@zv5Il{bwf@2-RF$rx4QOziFWm8LvetZpSP0VUxI z#L~c9XR|6**uqGg$ka)C*h1W1#LXt1IMgqi7Yd6J>OI#tYo;ZJsz^kzrkL54osVC# z=@n*sn(?pL9x5UrfYG6V#^L_9*)0(;5H2+Cv8ggNA22LQ6$68E3$i~Id2?zIomit; zS^&}?0HtX$?ksg`*yu*W??XFT`^`B~U1|A-EcSj-K^+#`^>KZ*W~suky@6>sLVaf)evhyQxU3X#+%xU7Q9`$XHz# zU!YXsiJqD~gv5`vH*3TWA9dVlQJcD`Ufj2W#m@UyETPY8DB4>gbc{*lD8>>_?L6_c z8y0xEER`UpJzbdVf<9*DrLb*47&+r?D5qduc3+{0oq(PEx(;7s(a+!(A(*Q z@WSBTuzH4K+?b?uBPTd6DRTq3k@CX0GW843PVZqkz%LL&#Wh{sMRI;EAIrs!v$PZq zKhxV5?vsK_?p6%^ySL12k`UPw({N`exy?$dKQuEiTM`^o(oMj1`q^S~IG03~cX#Qb2Hx1YDkxXZp~ zG>VZa`_PhzXBZrb0xb-v(8M$C<7P>jxEq6diE?MmQ0EKz@;tW^hr>EJ&zl|%*WGg2 z8{!hg)BQywxQYWvI_@JL$ZAk$E3PRvqVcC7W>6Xvf^+;;Tam)h<9f7XqL-q`*hU&4 zBnLn5lqZCYbYUFFC~3Tv(_!WZ-g03u$?^uxW`%b?kU<<`O(6kLKvRodVOBU7YCG-+ z)9^RF_YC{Bw)J9HlX0ylj{oTLM!bg0z^lj-x{T_`-D`ZMk~bp*vqKTJ_3(FnmaWTE z!YG@xQ9g)AN>ywZo(<(HOmMZt{76C%$Nm?AsIP8hI@*D7@0x@c)0HNf*y8qGGPP_! zzRnxcdJtt@OeHB`{>=GLt~2%h5EzHr(32;>L>C?WE(=V#;tt|H18FhzIV_9O-$y29 z&}Nj6Wlt$s;pm!BxOd<}&0u)xhV8+yb1*o(AiZp**6_XE^rtLCtja?bwfBx9C@+-;%Bvb_=uDKX=I$q@eRuWg;_@-tW!aN!icw3JG^teJW-XhRm^9* zO=q-35v8RpzjGcC!W3Q+wRPnXS3+z!#!Y;0N4CUM05t+O(+D-QjJfobe4#g2!45@N z&U(6YxMG>qSQ+YvD*f39??`l(89*4J#Q$#Qf`k+H=Dm2`MLcZ3Z`^@Xa-Lksez*Hc zd!-`c`W_X4^ll!Ewtoqdi~<4uE;x#KBxR^R!n@OGBp= z!Dh;R$N6NNFS{vk+MiFNB_?4I#;HL5UI1$6%NjAOwxkgIU>v4jTSXw3CubuL^e68HpmRufG^FrU0mX6=!$*x7k2iR*4u2p~VHO|;1-pNK% zd8MeajH729k`j-(5uE5tG3ZY@El;&EJgxZ3+ri81Zuk6_DV?6ln&@6#8NN_kg?!Ke z`F)5=c(i-5MpSSHt02R zP*j7`^;3qT!7y*V#w;aof6&_xhcAgFHg4w5PdRQTH|VW!i0f&@7kSD$yB+?SUFYA; znorImWx*DE(?M_>1DYP?Q_t-!Mk#tQ4vZ-4tanX^gNss&Q~q7!d`?*?Ty z=y?I5uN12DBJ@zKQZR17PCZ5Mv7OmvMo_>N!BI7NuRh8R5C~lj zV*3z~S+#ihFT*JtO#k4N-9Ot>u?!eiW2{>uOX~-=L=>v0YLqOW872Q-ze<_R)erL~ zbYsa)@>V0_qBoJ1LG%08v8K)uK{?X-ZZ+7$n<_kxqy>opQFpEc#jX$V zgW8GjSbDk&{6$?*{lK3m3G}R}>0E1L#)UBrV;i3$$)Djkx^j5}oK9BLATLy53NhNy z4=QbONB;Dori{k8Bp58rVr?b{QE^A(-$~aOmT5a{^B>X3uvI}N+fm9ohz~Kf8A{We zgL~Lrk!N2zLxgFlUtW-a?-iL*rUbpXF3Kch_v&b+KWed9lZ1^jV$aTEJiH3Y5@aZc zg!tmMls2Bnxg_jMtRy`CXOE$#5PAwl7>`HbF5$3z}n z-h&Dh!G#1ktwtWN&(9cc=egZ~Ui;U(*SIu{ygXmW59P17x?6n!-e9WR!j}H(aHVgd z$=%NX_IN_4_4T@Tr9-cO>u7MCZ^ouFRCRe)I%K`1DPKvZ@dBFy8;(=Oeq;V};Ar~| z#`1@SsO@j!j84uF$Hj3E%D-J5Shg;H_ne=pO!e#|StS()iZV_w?TqN*zoDggz}mg` zjX7Fa(>i0~8oR(Y;K)A*r4>`Xg$cAm4*|Tgz{pShTEQ!8L3v1Za{2SNwyarI~vwFcw&V z_r&I@rKsT>X!}AMB{7}~B{X6R{s+Pj#2-jMkbnH|9Q6m<5A+`xKQMn_{lNZ#^8@z> z-VgjA1V0FW5d9$jLGpw22iXtu9~3_*e^C9P{z3DD_6OY$`sRo!hQk_Ab}qL6??@}s zk#^qTOz3*9R+uoAMp8=$-S&2F-(<--xjMDAc1PfI&MT)3w^mCIsi>=XeGK#gf{PYWc8s~Zt+SoF&USG@T~*ayd0uDp zs@iuut9pJi@bmob_Km#!d~#*AxfEc1`R(X#i=AoJt&a0jQx7=a`g(``&a>E-P2%lI zE%j(|0>pV!V&eEEdzM&VE}%LF=NVSsy#?7Cg(BRS%o zXF2dFoS?%vk{b5b-)-2B1p#;j!8Wz!Z+z&7j>~N8@=uI+zU+4Xfl&_{x*hWU-rbyj z`i4jgaJhdjlpx$bCy10nG6Lk-p-87gCBFfwTXFP}8eXT5$T6l?+lYEL`lH8s6HYQz zN{8BKeI2#=O`(H-nc`F+kyF=u%0IlrkM&C7#`jrJmrBxl$ zB~n1eZ|3Cu7O1BGAhOPwM9{E<$c2}fI1Vo+SxL2Avys&t^;waW<^rhus@i0hQ4^^* z(Uq-zRikXRm4+{yaL_m@+?c z)dEUC#~m~ss^(}{Q#{e+Qr7T>?4Ag!&jVVZOLLnrWUg%A>b2&le>5`&qi7AH% zq}rWr!u17!NGCXcuK)l-L2;eAl#NBpAP+lEw(R|$$@LfMj1IXEu^m8f1sQ0@>8du% zJMStbE-+EFu* z%opV9QB70}33MFRwy7;t!JNM@yo}a6P~yPQGtrv`GVJZD=>Ux`4M?`xao4Ond6y;c z=3_2x1>4WVJn0B0(k=DjeJ!q&rY6H!;}IRCy|x%O3Wmhvl`89WTZO%PW5nL2t4jjw z)HrwbsI9YfcJnyK)H-E}umb3F%;d_23{y6m_V>i8>7SmX@qXxG_q_8C+E~v#6{%cm zTCD38Of0)CeSkxk?D6zkM55^o1G78KVW}obPL_DCntg;q=8iAuOoVXE=H&zQO!5jT z$J8~?btKmWP6r#P;$FAT)aPl1&G6=9b3yzWg=@)w4O#PU&e7tgt$!!QfW0^!3$Sxm zWJ-Vi8UeuYRtAc-gV@YDNA7@U1WINw4QTI%GzzA;DgkV&*p;eFK@5PX2BsSn*j(HC zT+^l(#o~Z?p;7*pVzpd~OKprM@a;X-g0;tUWMc>BGW@I0Up2!gzM0~0>HwS0fRF11 zN8C$ydXAyC`R;Nst;2uvyeM?u*SjqBmfGYhGrTxS9W9muD%*#hkrY6;DI!+oV&1IUXCLs42_>|cuoEjY6b{6V$U>?h4R)dj zFPJvIiWcf|R6)~%w`!)4T;`Z}5;-tS_ERh;#{pJsUTp-$^W|ldI$-|&E5c9$TyR>OvFs{PAnx{tOB#f|8gq%kL=OoV*`prQ0$H2 z1Z;Wc?nQtJ3ddoa?=*A__MzN(sLfA|wM=p4@$bEJC~>&yz6(*3IAjs)9DxV!G4R~n zu{Zz-V0{h=Mgro9=0SEll2y+dqxUunn9mP8(;*TV5t-vNwj_fw(N8@Wo3~250>g5J zcXrh|BavqCSVFPG7HVy$tY*PC(jtaV3!ad@>{K*15{vZ zwsw)_s0lO)CnSC*Cx05n|E=g-Xt6Mep0~A7ZjTC*Wi%SOAo~sG2yk zP;Bwtz!DY%X~M*#?&70NEdfg?;xC+v8E57jC{ELF{+XZrptQ9{tdjALuHvT2ctPiR zo@~+*vLZP>xsK{k0qv0TC)vniMTLnHiN_-nb)=5Cq)bG^EQ)AkiZt81Lj`DGAXhfR6#E@~vhezV9v2ge^-K1AGb^TF_zTyH(tmewB z(wpiH=*k-4)IpG3B+x@2_Y!f){w_@Y6Cbs3($zFaaZ*jDlz4cP*?@Z@B1mZvQ;0ax zctlJdCV3G`QSgM=F(10h{NMQwN8}z^oSEQw3Y9E9Qq>^op&O0JHUut}17lurSK0|k=I2kb;br*kfICcYf!+sMvv5y%-w$ErWU3eSfu2B=gSC_F zA~3fjLh@U?>~iMtV}RZ%U6vLcS+-er9p>-(Wkh3vGeKCbG=`8;$9_R)Pm~jD&&v#( zQI6F(+o7ZWaI(0&YD|zRbH9OSgZzcqx=$WEspJp5EYk11kg&edi}|?0TBNCYJmW~C z=Xx-t!u`-PldB1DlMopwZMU6GM-344Ij6vHH0S9aH90kx5da+!0le?yCoFT@F0K5O z6Xy)X!nGqrJc6aGv<7Cn@^xoYUAg-KKd)?-adg2Thos57@Ddw$0l|w~msJm& zO%>|LM>zjuqC9uyUu?%|{yJH5^8yw39SqI*h4RkUW}R5`cmPIdq$Rz3Ou&h*W*SBc zgE&m=SX^_#1b~IDjI+4M(_p3|TkdnMdCa<7zp$S9Bfdj&Y&q=sR6ksC%}& zzMSV5sHZn{sz$xpNDhl^!J`bFL1xCzYUhy@WF^($7eFm$^U`C&eOkqQC&{+2k}+77 zQvnn!`0IJ4TXH4)Hso(|Z-o_U4V&7cz~D|tTc60s+c0sr)V!@SDtMMclVYM>Vne=I zq^*~&x1r=~Pq-Z_id1j$8&G;GdHh!~kv1zo2#@r@Vw&wiTy45^<$NDPRnMJ4fvWlX5 z^F7^`^j`2P@Un^fLa#UZszRJ=gmK20B!kVRvVy9GTjVcW@d|W8{%{z(kHVeWQm5v( z=a^L2fj1e|H_<6J$`!7J$_PIbtZ?ppZj12F0)Q8>2%DokDi_jRIAW!{Y(!tv`SI5< zOMmlL>lxaT$c0**a~i%Vb-beir>8GV9T3jJLtssKU{oFkqR38Fn5YmUM08Z(WM9`b z!q#1xdGD#AO@osOnlq54sGX)i=3-$$>jieh9?6S;#-rW?yUHi3s)+x&+|GJlo8Qv* z7eJz0IjR$5b*_+qa$-26b6tIU3rp*8Y$h`t$xdWK*GCp`VPx!yJeN5j*XL0G*O$x7 zK_f28h3-%hczFHjUj?KiaURPb1X|MQg*~-bpV_Zdasw!lmSeR)5xb5Sp3VsT-PEoU z1#;+e5{HxU2bE-nePuAx-yAO(*JwFnfdJ?QpIbI7SVamP|G=EL;d@d02300ZOu+vTYU@zK zX)&PRuAbTfz7V%=qpH-3dE}NC5}DL z>#s->sX&K#6YaH(iKeM@M#{;tp45Oj{Is~~4$g1OZ@v5N#JBYb7X8voZIQnZmu%k= z3!S6Yvv!(foZ@3TqN(P8{yji^ zJ#i!37*mio&i_%igGuZ{3~k8bHvx##TAZ;8=X~o0r)7gV=t(Hnr)S~ol_+`!Xs#px z$9nyWRZ7g?%J{RwjSmYo!@-tIZU&FZOVVzpaP+U#*c}^khrdA6%bMzSR>4yeUjF3Hi5!{PAW;!*(1H7-iIIfYxE|2>L~uWVzqf%LPFESH*qF zb417%3Z_3&&&b*$HzZNG?jrw{5y)Iw?}UQ6K0_D>+t(ywvHYdR^;R~=IJz}jwk$xOg@c_Y7fgu?_(1=`L?%v`_lw;T05AYgR zxkm?fQ2zU#8Xj90f%bPV?HDV=8Qf#4>eAG-A1QgtZC9X)MNf5sfuYVjysuSr{kBR} zp120GW&#oM0pnwkD#Y;XPDrB$6-0kT=pk@ zk?FpCZfug+e!w|3KwPAwMIg8(TxkAR_Z4fTGw9gsV4)E+!NdxTG&4DN*58g>a70VX zM;9PXC6mRFcJSZX(9aQ3SMtD)bpw^)86WVmV}$9MV*R9{!K-E_;}MDpi81+jV3g#>sZeA!u%?-Me!(Tnwgzx|pjz&=ENy|COi!Tx- ziUU`_TD1WZq6PY#^?SO=+10x2TIoFQ%9dgFm!(+>XX2LG=jUAZ|1iXpm5Qhsh{=q1 zmpr>OU=tldO+ z|FyVZg2GZiz^ehcA?>>B_|}%p%yM+g91siMM-X+znH^uQIuq%flzL~M8D}8KzswH^ z=s##-h{G6=BPiEZM8T_b&%$U+x0AH$NxI`J6>4oAYj`bjpQO%_+an84YEOtEODkKXS~0gXRl__`{)0J9{Sk*RiX3- z`_~uhCFCVg%SHA{POzV1CJWV-r)l74*rx`jAdEi7G3GnUO8?Yg9JOn*6z_Myv7VyJ zpbbE&>7@YHa#|R37H8xR4NQ-lg9#3BFY(RAA_|?pT(x%vMSc3(BeNT(mE9pPVj3c7 zNai|DFPaoXBJdRWwdFAsJ%UlRt0}4F9H2@0{^H&E>kZI^UOf=jB+nBp)YGnQd0H*N zHiD=IhLjAawMzG35G2~F25tI(y2N2#*gt{&{WJrpW*L{7a#mQ63CfNw;b~pA-}ku0 zdoS*`J`}E1kqrTYI*SC+H`HLU7pYl|56n%XQh$nPHAGEUnR z?w}kX_^LJ0QA?}?W9~kA07+9XeA*Z8G+7+W)@p#8>wW#;2NGn`{Aq28eGEz;_{o77 zIWU_Nxha*B#sP1zQ$@jerEVy`Y<00-{yYbecW=oaiFujyev*O|Y^v9`FLuM5uYIOT zen-g6R?Utnv&{{Mvq)k3(OU)*8~^Z|ffU+{L$%3 zH>f;veDX7WTtw_0BSZ=!hSxq&+7EncV+ZqAWwkG z4y}sav|frhl1FN)U1E$H*M1*`ImDUt5jmJ!BCO^0Hbj zUW>+W&_y_1`5DlsLDPPth|{98L)W~|h%tO|)!`4Jwl4Z$MQ3h1E9@U+H6eakgm}Oo znW&x9Z8uo^p*TqHIy;h!W89>Mxfa8^z98`uyL*$%u`F8Gnk66l1dB{#`%~0JvBP-d z;ciPS8}%?Jr}mcfzR%bBH2eqfJeUHEA2>SvsdXan&6_iowTODq=C}P+2a%19|6H;; zXcTagOn{1;2og%SiOK+nsjlbKwDGD4Jtm`R0vCExs% zcH*#C9cRjT1-^Huh#$o?SZ?==Texl zC@U0&7L00}-`v<==%yctpWK!9?~dZ2nC#`A5J;YNvX^daNa5N*B|U_Xm+$QLM}&p8XO3_6!VB8pFtzS>Bfh7_BX9^=e zf8XG?tAht|BGiQeZ!d5MLt!3mxxt2s7|+C$B!Eh^RTAbeM_DcRhtsXHsZf3`QMGU6 zOEwLGRBvQ0YhE(A-}*^GIwG8qY^dlI+zFOVKnE&Xb=S8(3__RY3nit{_Vud z8HAFKdxb*Xxh_!Zg_5N?s#x?OUgrcF;{Q6u#y$@XqTHh;)6OlNN);OP=V^de*5oUt zD$3<9C!dZ=w(4vwMO<|^^LU5)fpLBcBf?^QBOc(v`x^MOqrbaXj;?I0HGbc+Pr3XI zb`ZQA5mMh)YkdOfqh4@?5sC3ciXJYIy>92V4mgP%57i(w^Y>0C0->5v<$i;l*2gtO zy)YnEHkz^uTYq2U#I~SCB|D&WR^w^K!8mWcZWuJudz&fC%4W&witv8@nSdlkkvoG3CI`I`IVzoZsZ*?S(l~c29=T+- zK!1_pJm*m#1=aJU6CY#JOcyEAoD_T78#N0(aLb^5mRMAcMKadK;kpOTz{mIZmPj}h zaz_R&9rpoF?Y@?f%{4Oqb;R;(v9=|KU5lR{rDAMoMG8Z>==E0M8QS9RwoC6Z!yfK4 z&8WJtd}}F4L5y31cwPu>RJ8tsd#5pAmDgwHFB1r6b;{3pEUgRR>Bz{p6To3wr_QBO zzG5BBH0I$|rtF+^9y*7WOV7G-kHtpyX?kGS%%%hs-z8}qj}`SL-r-Sv5^LI5#;wVT6Rw$qmDm40<-E)O9#G zs>=a3!bGbpF5LJOxTK|7#j(q2C{*9RBUPs6Y_yA8D<|M0viVWHN2R_yys~Q`8?zRPR>Q` z*;0Uy*_#;oFex5J`Dq?j|GMJ_&U^(FEc=-k>8ns)y^5-V-ElkAj_dZPcN5am1agg0 zlW~FuJ#$ze)Akv^K@Tlw{?;ui-e@tl+QWim*;O@}UTeCcsF|C23hmg-P{}JZ4&E<1 zk$}CP*ibokC3Uk9(<%Z@58ox*{NQ-FD(FT7>3aOSb$_=|>d^HwGRRO7>&(=C6}FO3 z^4UCosZfNAYNK4P!XLWh0Wx8`e2ho+wtKF}RXGYf9C-z0BN7?(KJ+K7Lu6TpJoua%*R+_Q~1l zR_e-c|32FpGqXctVmx1!xxp4?Dd<6U9 zL$DM}K4eRf*zcG1Y;EUWSZaD(!CyI!t<8IXm#d>Jn*3T!iICVr;*m{%^{1J=k|q$U zUUQZlO|>*sMa9rwXL8zlz9j%;`hm#BN#1n2OE3S^;wQVb>zA3kZwB9~T}RBY3A)ig zW?(5~+?-hN_X&{ae8)?^@DOlU`9^s5if36>UR$|03>@8X1hR1QnBH_2%&_~N(>g(k zF2|CBaO`~AELv+EwT(8L963Uefhs*8LmAvF&vk|;m z#bK&eZCATqOTm2V_!~s9KRjrTx&PqQ%We&49#dZB;}wvD9T;om+LK?`_n9*+f4E`B zKg%1PyB-lq5!PsH z7dvqmk))+tz@oG1;#>`8n^mmwH;9|5O~6p>BL#en88l_hnJRbcS<>?SKGsh24+Y-d z>>AsYc1z^h?XIjpDju6)XSkTWx;cwgU+-YYLQ|+tOFnMxt1}AC<4E8hGv4^8Q~Q)VFE}QXd4wZG@pgNi=Ot%qYqQhYY>=;;C9iMh zul33Gxr|q?W;AO1(Q5zsbMYjP0-VPztOfv@KCKNMP4fG_)6TQyIOO2t@I60VGVcdV zF!$cL9tj~1cMKK67Fw%qY%m&)gFb&$S>lT$@T)H@Hi>fwn#A02iRB~ZFr~RW^H%0u zBqIR*rj;TDc`BRclZ=S!FD*lhE3i=n=%sOU^b!wY4HA~_B zx!V;8+O9FY=vR=yA1u3$f?>rWUq3YvBdpRs$+uKnL<=6`R3m<(v)VZxlzS7pe-pAR zY0T4$h;OqD9LX>fsEXK*^YH){K6L?!$PYXB;)UCQ%=SBc+OAVGbKa5<*r zA+aQJ^QO+k2;>~gZx0O542^kr{-1}*^6s1ZFFgf!(j*;;_oz?X(bn?84*K*ROL;%T zSAM5#f02)Fl z=NAl(@(GqgtB?qOz@^1|+VyN+%T=8~S81nb%-r!Z}zV+urtxD+U{HwAIY4 zdktF*S$M+>nu_no^C3yN*>h$~2d(7dQVa8)s%=eZo#uXz43V#pP;_!Fzst)o z61h73twiaR-W*D%J@wDPfpgAJiowqfieixYj$8R=sGi?d)T{=`V3N;eWNG!>9_K=9t!2t{)At4mLTwfu81ws-dHu=Io-? zpB48YC?{)I`lzWV%U>B3cxh1Q$~MvI1eBK9=B$>FnRiiZ-uo2PN-7|mQNN+~ex&$? zOwQAyQN#YlQXB^WOZ&t#o0ZvWqRQd0tmm_`Ch9lT9qv@k6}8pd5S#emPw5Qw6liWf zZ@w~D!MH0GcLPaTX1_&)`MwB|!I!j;Z#6net%<)G$2Jn=m|5kVd+BHo2jO#3G}O!f zbBE$PNhN&Y%V$wt>$5tWU>ObP=b@f7*Pj>*_=J*=G=u<{P@~4k(#}Mn`CO4f71r)~6p{^};2q$fWL*R1OPSxZ%T$tb1E*m;|1Vp1PscZjmJiiarFI zN*E^Y@vwZ3H7OY)(t2s4e-V?SVQtXF9^q^S@eGP^YbLc@bAw9x?(4eF%b7eGDoFc^ zt>(fk{7w|bSi#tHR0u_0LoF0H=$~lBr@p80gPQ@A03Zewlghurd*+ql^llxYj{(<_!EUpBEfqQ20L7Z$MZIk~EBs8pFq9{TPHEy}jqp0_202xPSqp{bk|RH$}N zGMfQjef}pfX}#`24h%v^oi#UU&ehQU5k;OUBag7XY0J!d@kt=hI$1I|)Fd#p$UQMyZ{QY= z%EYq7Ut)C5@*m(DQrlqd1^yuZJyGjhrz# z;>f7lWXH9Xmf2{dsRHi+Pi$m|%J--LJ=DE^@Dlm=^4gW@CH6Pr;irfMKlthW6dSlw z&rd+MbB@yKr4VYYrjoRhS3o!Z8vLi>GvpGKgsW#3-Jgs!4QT9@5mQ*KMWC3(B23KE zJ08DG2ev*u`jbBdC5&T63lh&KOQ{r_`)b72sx-lZNx;l-xf7O=19vc*&zw_7^ETk)(-GvptdbQ8N*Kw zi_PiY{4$A>nw_UO*>So9Exq==hIq18=g;&iMqhK)rCft_yBEZ^0S70ujz2Qnzvqq~o;GO$_k%ZX{l?f0W7P*q^g%0z5c0H8ZEhwi&Lv7QfrjDs{1c>L z>b-1&*rd3I`*!f8bd8azP)Ae%dr^r80Rnmp!Kk(4paoqDM|^9x$oWbZWtTD&u-!L} zIC;&AyO=rYCjYiwpev^nQH2`CX<1s^z!m8vE;iHWgpS3ZoT`J8d#&nhf<= z`R@sqgYUdJ0P@@Ye;bocv&XuPO5PEfX$Xs^H;LLp2F-b~e=hZW-(rwEK{!2~@B@E? zcSLq4*aB^@>g3e00jp~R#Ds;i*8wvq?&Ho)y*$nohq8@yOp3QaPKH2xMd6d}t5L|E z-jAPb2>%plTQhD~xMx}Wyl_v131T)#LjsGs;PS9Wuo8M~!KZT_X}tJ!WnwLYIf`Oi z{)muU^h;0dSUY9FjS)u@!}8@|T|K`B8tHkX8Qh6tnV;4)N#F=kL_Q59cXRpr1f|`v#vzRP!0ygRGEa~Rft4<>bcjJ6 zN11EKv6xG8jOTXxaTdl!c(Jsf>K8er5Pk9@xOL*uDQeXTO#Wc@{N3vxfuyxE)sR@sf3BH8~NW&IKK%}|l2WdMwwIg@GjBt!`jmMv;wn!L(kMaFfM|0=LGzp5_ z#eB2pyqHQ<`9xR?zPXCPwX%p*CLvoU&W;Nn*nDLvQ-d;rmlRCq_rOJ(_}8heQ?Za~ z2oRfZwL(4e?IsK$9&>UmEB+1bIZaD9hw7(U9{EE#K#Z#k0Qd14l316{+HhW>h(XI# z<$LH;Zc9HHH&?9kkkhQrpmH&=S;K6f6wM4gsoh&WsqHrv%KRAL>!Xf=VX1Fngpl3# z^ThNx$h_WB_yR5!g@h78Hm)k6|yuzM1?AN z)yS7TczikH=6_k<-4@<6hSi=tq@sb%dcHa)zy859QS_;X7-pUJ}2_@6EHap=x z5A-ubx0um5yTDw z<^0}9p96AyqFu%DKIrbAEfMCQUMh!H{_(FlbBgrHvwz5KUFMc|`>2AH%uP8YHwSNP z3u|c_$+frW2Xx>5=}yQDYbTZK%R(~O>}<~1s!9Hfoh;gDqQ@g5j5SSw{1=#(~CN1a(km3?4HF3H9O>b2lGO4xM8)kuxuz<3Nw97Qo zNRc>wZ7ZbIVv-6AZ8FZ27Ufve$t*&ne5a=&f4~{4)B7nrU{GSv&}9z7W)<&rX)u!~ z#625gkATPJ0FO-Ys&t7)kSZ}Ey4VqR&0)Y>42R#5m9NOt%a1H^HRkZIq}wwPV8HW& zMf>z1knHx$J9{`0h6w7BQ-UwlA6K!E8AP7!J5l3MO1sz`<)q!#fwfjP3~H-dL4~=RG?s3cG3*>>U(K`Z1ok z>_!f-Oq0Xwk?q<&ID1%rk;Qe4?4wmuzM7D0Q|jkvk(#!c;#;Lda6B%fg>}PII-Fy= zL%uGkP!XB76n|5zLeAiR{XaJce0AfA`9&sZ< zVi@}KfT7)TS}?Kngq4#|c3e?5sWlRjZ+qQx@16VV{*0SL6ZnMRdmL@*+Snq`w(pVQ zG6o8p7+o_=+rUT;eUn;veNnK9h!RuAxM5jQU|poBDuhS`{F>P#YsDyI{a6>j*IwlD z_zkjTEa!M!IWYVzPw)8{ut%3Eg??1+8H{&;pgxZ*5CiRTvK*UHFR+wMfK3QTF~5A7 zW)Wf2$}sPodK}TFCANm<+Db~^SmW2AWQUbkQU@HafuvK_u$kcdQUGZIqnElQxyi+m zuty6e6;FnZq1ey_*^O;b`8^9{ZLa?P79U*}_xNY-8_re~qgng|uo^0zFO6P~*bH(> zq!5SKRnWi#*EVeEe0elv>dTv6$}O3t01}|3QadVpz2zC#2CRdvb!>23e%NIbXx?0yLwzon=WC-~OryKL?tcFpxCTlEH$!&YZrEm2!{6F|Wjpp|xX=YWvFpmjD0!?E2# z2TvIRx_Hv4P@gfZDx~FvfkJ4YL9@0t?aN~18=2%QUO44c$)Roh^FRv}3s@?%vADFm z|5WW+K%4E~J}{Ywb~y#QzplSacnhY>l<4|m!#uV%8s7-uJHF+|q&AS0n>Z8-d8jd2 z8~E5BW>IKQ0QTb4kCf&cfv%~X!hbT)iEe!MJBG>;k;C+OChp;E%UbLWM6)S_q(QM_ z8)|RiAW}EqrHQxlF53SkfZJA4N(rU69eTdiS1LdvBrDrqSVo|Mjg7}b5&;++PWm+W z2(GCfVxI*-#yokIK=a-4IC^oZqn#iE!wmB>!nKh*0O4Or=J1P5j0^3-WR<$2NU(3u zSfH~ibj&@PN%eJF&ZPRps9sH4)v1uh5siW`#e}1R#3j!#ajiY$#gvC7dJm3BSN(<- z68MU=Fv6N8Sia?eXG2IwO((Xd8t3ASflh?jP8NnW3pj>R^;2AWfy1RZGF3x$eD?as3PKZ(h|-SOn!%n7t?Th z1`fe83!qr8(R+We#mh6ZCTQ1lTdfmQt+2eO)m-8h>@^A?El(yCLx-h%FC z>3c1GqG#pTV^gR8s6Bz!(3?cb7rbC*tl}mn%wKl`$DRflt=XRn;?$fU?d$5_5Nsgi zOuqqsb*p8d}qO-ZvKFY~H0<=y=9G z3dE-&5z`y{Eo?Y``AltX5;E>5MdIf@AJT4X=9xHlOP~V)$s<1x0(N2 zE;xua&Xp{0zuaWFxU6)`&$w~;eY$#o-L2OlNQ67YO%uwtFf2|hGsdD_741qlWyc|_b<2?q5Sl(1%KX>WK+@hwM}QP60;Kd6fm{Z2346AoP#AMwOp@ojtwj(wFE z+8!P=dfe}257@{ zc=gAtbyhcZP7Q)bs;cUIT{(J(vs-^>PODzOKU9l1r&s?8|LHV3$4Vwy1+!dBnx8Tbc8Kmk=rAX?liM@~*-}B9olmezLV}V{S|3gG-Piy$> zr=VT#{}=xML9WaeGH_QEicf!oO1J|T>KVXea0aT4@Cz;`t)tZ%>@hXF=qO4LTX-W2VB6%LA&$S3K1XD@`sq_Zt(K zvRoA#G%QtKi&~l{r)gNFWhJUKlS-M(mjfR%j&|UZPHIAl3aI+j2?*Yf9`wPca#gEV1z1-P=?Yg_CD`g?q)2WWi@g{+8FLDUst!H^=_#`0yfbzMG zNppB^b$g08u16;dKZzJN1t(*LR|c=NxBD9}pse%cwrE5>_(3$@d!Z!IX2haOZ1CRx zlQ=;En4UA#cnwv`8PIHogvm!1TQ9tZ2k;M!#lyprxKwrLqBL@Kp3Y%~c~M31Hc z3LB6-3*vi4Z|U;y3N71jbf~b+-;*r6DL^tp5l@Z{-<(um91WMzDRur4VMFzapM(3r zSHE_W%&rO~bJ|qvvM68+wDkXiW(YQZdP@(m`rjtR`AU05+wOncyS8s->TSzf=&oma zA~hErBEFIe=mlJp#G@|9Yj|SRjos*O;2Z%m^iSO7_rxVo`Zu`M2XEG8sb~;t zxZ^9w``rtfVDW!B>bhrwpXwT@MeF#*&va+h!tawBa!!?bg>fZ12%g*#^z2lqw*wdB z1*n+lcav6&fcx)Ip*p(H*7J!pJWA<6tqP*=osz-KaC7*RWY}j#mdEGTgcJia=BY>P zXSCxAAv=Oz@SdZQM}PZPRNhg@8QKo4?i4#bM;z*)sAVH22jiwrL`}jeQ|kJf~}} zl`yK)+!YK81S@fOT$rm37OAb*?tGnY{Gu!P8n>u|HAp>_It7{dxK+A6j>nK?v2ezw zcg%`gwNZb9)4`^vE0{5HTOh1z7Em5p4Z<~8db;!0&^%aW?El5k<zkU(>W5ok2A zihj~-=)@=u^uZJbs-NS`P43iS?nrsG(-sk|xmbT!g#!~PTW7hf>(+R$nbquhQj2RR zHhL5DL*V0Q#d+FAZM&q2$RChQNU5|2c`i5vgZu4>>riiWqP! zcpbEk$_UJ!xk&vC@R*|AT5yw1;65{>o@_>_Bv~_T)g|&^?lPHX>ReegPH>oUDbUBS zq(y(tsdkNT3NXRTOyNz7^Z@XUsAtMh&5E_l3=IV=rd}DRUKt?D(0m{seKK?I5aAL! zJ_9{d?u--!1?wSi+UupMqNNKifvM{9%88_QCTOttrj^k4XTKa&B!uk(=N^^)BxMMx zJ7SPOC!9mrlE7N4HN5hHt3KR#Z-aVCp%{NRkBh;1ah`W#;r^-sZG5qGEBA^bI`tvp zS>~19K*6SBo_qeuOCw^{^XEvtLia%ycu+M$LO8Evv-EgY4$K+w6LRQcHsir);<@-^^T81j7)qu>Z2@?gyp1#Lj=9|t{FX=A|30$gWSh(_D(7?_E;Dkw6~F*+CA9W zgS3|vS(Y^GG_6&UHHmYg$n0@Oku@(WvS93MMOK1Qk%DcJuIoI1Mv>K`$ZpYKV7gL( z791Y?c#y7&#L&U$MCWDY2x*YIq}zYat{gacr?BQ6{e2N~Cd?Nqv1vsKrchoL(~7dF z%H@w8nqow>NX%8Wd^30JW?ewH@73DdtH%owBi0C||M&(e`Fl^UF!CY9siu3e)5nx$ z1wMOcb)+PD>7m1>aLUr~ST47xz}J59&i(=rX~@gaOSo5@U|6Cabr|P=t{s1wP&$m| zrmM^;>e%_!sW9L4T(4U8uo^Rkr*D_9tEOJ0MGp3FiKP4SQ?l4W9q=+RMZI{WmWM@} zHf!flL+%$JfDM?$L&Mo6>vECi)f!-NY&Qc7#CCHQ+09=9G4ntk?sWNDe%pZOH|Ey% zYOy8s%J$fi$6q7fKtwQ=w9|eZRa+J7GAWo^ zgiL;g%HPn>fRuQ{s`iIT4)IR*7+#e58W<9!R=##cqj<&(1M+lwdvTR#=Iq8tJ*hpv z?$oT?dRV3EbKoYAic!pF%b1)=>vw9*|EUTwizRrL21=zRxpnWWEZBdwnzMv9fjcNG zoU#Zl6h?ex_=bMEqL?s}u#4?rFJl}+rEIG=(-j5^opqr7qL`emNb7A;KEjAAn)1Jf zVc&0RLtm7F)u@x}x|ZC7S|Rct@0YzyOl_fTdK=>PqD;6OPC7`0pfnVUkAp(F2k}T_ zpFX}502mJkYl zI(Y0Pf8Cd z6A8vdt|6nJy;_N-M_=B{;=;X+e2e2r;%EYDNwijqoZ|P5`7(dLd%<^+0UP{3!a-SO z>^^+u_b``Yh$04pPw)6~T1P53BV$GetIX667WBy&+^gZqhdi;nk+c-9Cyk$azr4%H zUUI3(ylp5{xgLQT{`CVAC5*){HtCG0Pdw;ZvM34+{Fo!owJ~rl^xs+?!ETH*( znzSj#OVhGz2_i6YO1P#0Xf+>=P^z~PxiAd0Cs%?d^Ty^VT*2}eyP&(clvC+oofMU! znyFlgBmui23Iw;qtV~t8*fd5B>a^^&;EC(?j$r&sd>McL$~EIO7%U4<9bOR~CHxo} zb@1M^PWg7@>Il+f-5f^ghY6aVlm)Gzd4^t>OmuF!$j-|TYK}fH}o(z zKWjB&Vh#f_T9=8qSD+qa7*nwa4K}(Y_u-n2nbSM0$(Y_RvhF8t{So)_2@?|v&!A`k zVWpXUhi-p#SB~Nl>M`7(d4e!%iY+1s@{H_9c5B!RQS!CdvHfHDR-XlrVGC4P9WNK& zE6>W5t*|1ZTPcyZ+-B3C`Qh5&WL9jd2_M*wE6y*5Nx&-HJ}+6r)SAkYDe_7&DXW)W zs-KC9@p5584lG*$j22s292#5D0G9v*oaT}+P9cBN93_c~Fnj&Z6=1xr8#La)w7^m7 zb%u5Zjso`i4GasM{`T7qz^KjsLfJW-v$Hww_~GtyzJLi;b7BMF!NUW+p0E#fY%$Yk zn91ouQIx?FGRqbszISP48#@}`L@1`+$azmRv{aHaw=`FqOH0zcSa`M$fsI{(%#bTS zBNKm>X`p`Fz8zvUI44hOJW`yopMqk|S%$235$UeCpXIcJz*6F-0x{Ynzn+9~OduS6 z`T~!vNfSgt<`~Liu$twD>a+0?#=c<;#pMMGPc`*gL$Y}mu*tnTz#@MFp9H3XxC^jhUdGN&GBWdOYKj<4QL89v zrF<@-rXo6;if~fTASC|8%_HEvn%&Sm+ItDG^a^(GS?r>~w=5l<+^3lJSKd|B2#n4avp!R z05sIOg`5L{pWcgTGJ_M!xS~-3qNfZ3;87JWjcmhKoE#;rw+&>?j43hFoe~Q{``?y3h(kby z)uYs3rK(`*qa^l{M?sxE?4j_xyB~k2`;sRjvXshn?2owROdEt92al>|vac*@kn`*g zFEW%k0s2#ruuc4KIU?ezAg6pfIvy+tlBWqS4pGJcDAGc7BGE}xLgU0^9f4@boQR!Pk&F;A^MXQTFy`ne^WTCy!Z7>#+jQCj=3Nz+M=JGCxICti6H?MDy zJsvvbKdXV=&T`EDJ5Hs|Emv0hb>4Wy!R`Y01`XQ<{%?3! zP)VW-RrI!+l3zmWg{~PJ9t#Z?;8ah4M2vz{R15b zHpUZ=@n7L#%2IBfP{vKoI%#_9lyI{)~n zw(MT24nDgi0Knlu^17LGPEh4@zPl*LpMi$&T57qB848-MCV?JXxq zRZ&_`v+LNy1A?an>H3S$0mM0N`Z~_Ky-p1{ghoIBs|&aW&4JLc0+y+VU&Qf=GuSni zwC2y6BrIHH6WAdv)k%Li`?wFjVZxeg3#Z6xlT(va7X&Ind6kjJEG^Bs0FQ?aMeHQV zpH?Q@qEkHdxB=*7v{bH^eTs5h|ThdIML<_($$Up|yv=#eH7Q+T04$ zDDOePjleS{3_cikFz~@GEk7$+e(`xU9v#HAQ6%vfQwe4D3SR5NoWg{IJ6({>ZbOhy zItS{#n`DH*rt5!v7UEt-eD%x+f<&a>aYSbbByg_^BrsT;GS;zJV9F}Md7XS_!Z57W z5e0@pxuXys{xzUAqpu9EV=z>FFyUg~i0%osWf@SY_iz#`XV zY>Uh#fFRVT=t&J{@eWp%*2l;iob-s?cADqPm$a z;T-y2ckhi=i+F^2FmvB%;9$|YK0WIN8Q4$Sjy*grm!AV?L!6CXm9gu&asu8z;|qS= z>?w|r-8g@6oyWPp9>k1Y6-_fk z{Vqx7WUey;rv6unz7L}hl;=NHnz3449V&qk(KlX7m&tQd8d{QB;)ICCnN(MfFXapv z)?7)G#PhOfz$73K2H zbSlAvEf}`)L@V2|VXAno&LYJY<%lQ+t@KalU z#6|p@eyDml_=@4h=z4tl;%G$6ql~lLovS7}z&SU^0j@M%IPh!WSa;4vArsQoGX0KQ zPg{Q$e%ZSKTZZlbK9P__;AYl#Y5C%*w$2D(!*q{kAshb!BdWYP1 zTx;H@Ic&JK{rI#J|Jjm4tV{Umfx<0Z@=<@8G0Y%2SWq^0dr%k4cnQ#9^p$Z|#l}7+ zpp1|YmQ8cSdQzKPMtGPacMLpwXUh2YX%>Z-L24Nm!U%sU08kWj?1nY=?~)X;4PQ6~ zxb1!yF6f%E{^~cy`1fA#YZUaAGGehT%9TeRuDnn1zqsC0?{Q1E|# z_6zLb9|~!5|8Xn6i~=Q`)de-#2n;F0#2Werz7ZE)U;Q6Ft2-@+gwZm$gwZpdP69MAm(le#DSyRWTXWkumVWPF;o6sx zsw_6aJM*-TqBs}Z$=c3lXS(M>k!hQWE*?@!x6?bde|>?2bASU7@}+C4=0O%k0w92M z|IUHqgPVi+;P|H)zh7Sd^z-XHI}jqO^E^4Yygta&sI04lvaBMJ)CZSW2fvKUXgm>d z9FLC1c|MxHKY#l$&g0SNli91|)7jbO#rW@+?|y!rii0Yuiy~9QsH^f|A`&@pqDQ&B z9g9&jo}_s-n#;?b{A_Q>$!OE{`tEPzq#AAT(Ncb{^nG>Nej6vX?p}|!^2SC#QlcXFq>OFU=1>XMc)()y>HGfx=`L16o*LN_XCB;!z=oNj@ zBQi`EgDC^Qj646UefYb>K;sNs~f2Y9jqwE#R$C`vjJs^J4&ljJ5*LMT(^O{ zExc5Ye1E0eD}wdS^$vGjHg0ihcSLD@o6lJGyKWja%okkWLJj^+a$8|KnbZX0&_v&Ugf?2QGg4E3~Y9{x(7qM8Qf4&}9`KW8K z(?V^~t?qP2WK#UP!3LTWq32Y0QncuBuRq(??IN^!YQBkZ=Ua>}kLC4DUmqdis7KrD`%`wW-w@5) z9e;`h4(!aUVDAYKWUBUZrZy$g3eQ!XvUQS2af~ajjLv-Z>h0y*4`<^TrJ};BpjC zpGH}fmoy(ETzaSBk-O7m&|0t+bv^n>`tct-d|T5=jS#ymDI4mpEvg^+{B&$7y?>yY zw=hk3jzoIP=yR#)`{^U70@)^3yRN=26;-&srR%;q3`v026FwC)O-1Bw))3w{?goM*j>DP$WfE<+(#bBSm`uftsR@ zqdM{EIbE_DT@h`v(Mz%~9hsq6OMlHL(oV)F0NIqHELGsIagvTR0b<_xci7_~GY(25L_N8Py-G6vUpt6wT3tA3sORMPI+$-)?a`7NSMQp8X=2~zl+ zD9aqkPEFiO^yC8>l?Gj_q(=P&^FGoesNM}Ou15_v-*VT>_(x>y?|(pfCekgQbH7bK zVm_r=58DIoCTv#R>ssjkEC^}%xP5cu1r&p*B+Skt6;UD?3>iT2@NB01w?g`IW9i-l zKZDlkackrg$;?^zu+Nm1X<$dm&^6tmZ%^UpJ7oN0W*EMY z^U@e}!$d*mJ>}ts5so>&8I2WV;JJnkI{n}ZJ89$kBNSCamRE*$w08HmXN2C>+VTbMU`+7 zC!oi+F>JHYR=v@DprKTb>f-gkBWI-X^uA`u(sw%nW@$I9SRxF6s}*9 z=(k&ArL+y!uXC7fr7Wb`mxU$&f2o@U*N|4>iqCk_8h_*JdhP7X;qC{Ld$M6hnZpUP z*{AnUWVn5rG=ET&PD$lae7>Sx#0Dqxz}%WcyJ*yrFRg;T7;{akP|HX3>qq4x?boAf zXzgBY>`M&drgZ3rg)Hb80PNTA42c!m0&&_48ENDNIWwVVdsOWmIev<7G)JvM3W_sZ zZ^BH&lYiv7fOoClwdh)mEw}hWz?c==f;wO6_Ldb}hO#cLAJ*P*6tB0W0Y5zC>eOYF z#|9bTiZ!BuQnzr{M~sWU`)#8Y_}leXBLhOa*;@^aF{kAMmZuB=L516U{p6Y!u7^5x zHIWV`)@Ii!%4P%_qIl(I>gW!A{&BY^lRMMea=XFh2NwP{){pxg392?^2#d|sVS*elNCVr z@-P6^)|2WtcC_~znulzOh6xDv?&c`Rek+?3J^}4+Vck8Glu2MGL-U+U<;jUb(%;s3@&7^M)**P7nc& zRFiM|af?QTK^({>p*Ss#gq;HgU3C$aRqALs40rXL?=(bWc=$iYA}K~6x*KMTH)KN9 z&3oMfrEW&96tx?&^M>KBHynqX4>T|fA*?I50in7gXmra$g|+2?J02nR*7M17rhin4 zMNrF}nv9$_3;@n$@P8#VkXdRen9cZ$S2rApOhv&%= z&#O=JRL-t~oc$_d5U82CakJhj1afj>cL;JMJdJSZI9_o5hH=W8o0@@wKeb-i`34~) ziiVs0Zrl-JIU5F(Z<))cm9yH;FMkRv!aZFOF@b)qpJlci?R(YD9-fv8e#p?SPZi)1 z4h**&{W3_bE=(qg_CT>qj~lel_mb9L6e8?;X?-2?vItY9gu{An*B^AWviSbw$y!LxH$ zqCX(Hw~-rM33T5K;;EZ08m?lm2)gX7FStWFfb7@8a3*CwpDOP4gNE!F94B4tfgW^s zXdx;#=_}t~Ou&Ls3{5+-qe#J1cuc~l8fSBn;=>1poS|lI=Ok2?E(JVz4fYNTAs_>O zXaw2#UrXD~Mu5oU+74<4F@LIK)3L&i-f{KdoKB3j0{HBN#owXkf#*361X%M#F$7#c2`m9Oa7F4n<}b-7GzOc^F)W!xT_FJ0CIy; z28l_$rl1 z)Uw#ezsxFl*syr!Q1K;*F{Cy^AstDLE*02=?Vb$?CuPCGb7V^b>4UItc^ z`pX+)$CZ6i$o1n@Y`j#Ww|t7r>Tk(?LNnId!a6S|@+-=on5KDD`yL_h^^RW{^5lqn zFDR%WrN|Ye&1X{Y@05DqX-Npk8+0Nj#~lMfl>>gYy4y^ACFCgrd|BojS*fy(3>MP&`HY*eK)JrswzN=H;yq z`UMu1GU@q!Ed9!+aPEZF^G<84dWFq;*D#3m3 z%jc#=M|!8#vPt*0+7&aX>DJ2Ul~eZFeql8O#`N@i(|`SOl-Bv5PWS8H|Jro_3Bh@5 zAQ+}X;gJaxmE%ZmkLwWn)}u(GxXOI!iCzus2Jo8bd9)MS2u#~t+!H#j7$v(So zkqCuE#J|2oru!uFwBw19|^|L?`&CS^H`Ia$+lwUuVi6S&3fQhU< z=T2iHjR`)2wc4bkt{aLqcz6`3XUL*d2+t5Yz<;R?bu*Mv!HRh@W-vd&B)xhsQ^jV& z;cn^WbbC}JI2UI@T+-v`NP4zU(vLZC0r9nlzzAhK+OdqzipL!b1|m`*-amF)#SQBr zQ0`dCjyVg^4^hLA@RNGh_6e%`trVuQ0cy6au6wQ{15L+N_?B~$lIN)-Jl7Rn5Qodb zEq{DU6ULP{kY$!FerxsNrT~~=aS0OyqF39oP1EUVb>nwHa9i-T7L%taYRmfuBZ$2AOlotb`cEVp*q$TB@?pI(tG!zs z@f>cyH@*XoV~(}GtHY+v(!hmq(Cc@IyK<9f;NhArwtyvg#TEgj%;bNFOfVdy*QJ@P zek2GUlnOkk{!Mvbs%}y<7<#HU7rfA}?odk7l-G$M%jK+Wl#+6O`6%Ua|G@8BxPRf= z8EgdN+0oPd#5{>Ywxif5I^20{os%0JKM$wO_~DyyF6`XJ-TzedhIQEk@nK~Ty=R{K zEn{-5IBVJCJyp>^`=vyz9?3Hl`}1sO%KPhzB)?>-A5&o>TR(fgt2XtsNMQ)s7W+Rb z1Ot^bfPkHX>*v>?;>!q2THj3GlYfs#&!#5)4&us58j0GMI4GkB{k!x^fYM-!NknzX z$dr#s%EdSt0rs|8GXuhS^>TJX#|vk0I{4_tAWjK^W^S&71ta?`2LOO;?%A&V5T?H; z&#GB@c{!c~fheT;vKIx&qoSyun$p17U-ua}8ZOdi&ae1Ln+;_y2>$wWV1Ju!hVUAe z+2Z*Y&!_tu0Ir7&m{d|qLuUmZ!`T9in3Y)A<}+gTAR{JvW5A=_u(|}|#Y64Ap5lS4 z$!p}#1P=q7axn;^F$OgU&`U7F_?{C6;o0mXK7?E{WG*%HftUVbFUy{Tr(vx7g4MBX zuDJ;@VaA#O@wLro&O_L;?tjfI9LC7xEQ!J+6Q{z#bzTBWt!qd=bv1)2Ww0jBoc3Gm zY_82w>P&_e>?^AUw$@rqFHfmNSbB1f{gS?GN4s{%HzzRDH{^nDH>434Bnw!793X(R zb(iMw;ILkYxpyX?0Pi@oN2oMKS~_xZ7?8A&^v@N#EkniT?3mao{eR~Ic339CLz22$ z&dtqa-thxlt?%*J98Vhp+$}*KUg$wYE{1RQ!7W<^!=Qr&?!^&nS~hakbeaR5?0Cg&x*@gmCtAR zp9fIl?VtLj0YuR`$dif3lYg+|0)hJHGQ@&;78Pj{ zU_tR13o8B9eScwW$kP#97U&`sKl2j0rIRRh^^TU;cFoieb8uc#%H#)(6DISJHG`Uf zA2nP9hjpZHOsUQi)XNQ~1BWE4`9J5uD3oB>}RhdkF=OVWQUgYffpCVS%N z5ElIC|4ATw@&wlL&{T(u{NwvUV6`fEcKaB{vj3)KFDm-Z7?Ou7PU<8QwFodoQo7bA zb@fNiGUvbbMlq~Rq3_w{PyY|Wit=fP^EEVw^EEZM^EEcrR|7XSIhSCsG!+Ch zHZV4qfx9v&e|BY59NdyE5Fki!_r~4b-QAsFjW+J??ry=|U4lb!4-nkl2_F3MeRuBM zS@*s9GjD!$uRgU;?UG%mR+AGesnQFZ+8F~S>};LsnHiXP01B2i#x71Ob~Xxj+;a3P zKyw#qXCn~c17n9HCl_@D8aZ3q*@_uC191GIJip!gp+ODjkMs3aBC0g^ylpyNjoN-oAAOA~;cr3uj12}lhvvvUN1{&4`9*x8y| z{#BY2f5QituoJ)t;A9UpvHWlYx|;y)|AOcM_CQA)ODCs~-vCP|fVrcQt@B3@ob3RX zwk9AK)4vk+%&Z_arvvHzp)RR57gPn($)#!40Lz?e=C+T5MXNQWDhd(_=x>MX76bEcN#8E zmbT{q%76~w2sAfxGz9^joIYqixc}pyd_ban!Q%oyO9 zSw3Q#IDf=8x3q<0{A*sMZO!Ze%uN5Vo4VNl6LSSR{@oy|zh;K|BMBo@J6n(kz!Yc( zf5)g`=ll^8K=nUUnc;s={72UR!TIm-KSBuG zntv!n&&`ecfBUNW3Thoux`%lAPu!)PKsZZ-bA@|9SrV5deYiKohv- z1v?YoV5{`34Cgwm+@T3E-Tq$YPeUR0rHyJu1^zX5J$~uoTilg~FSQ{IO;bE?-#vbv z*uU<{(Cm-&VuC1pQh^uw_&%i6e|-fg>l7WFwEKaK2BioYr<&1Tm-BL)kUAVp8vEN5 z$|?pWFVF;29juNpmk-qbq^<@2=w$cyh`uIi4RIKq$W36Vb%iIV6iG)9_ml&p!$X5T zS5SR>c(gfju;jc(SiUJI#L3PchK{$5CT~zruADCLZU|mz8Luv+0k$ORe|Dnvp0kfE z{!|^MH$hW+B56*I%N9*co|i+n7NY#5fH3@!mrPFY0Es%%Sd-A}-^hFYymv@U_$eE- zGU}OKu3G+gjZ3Kr-V%&dteI!h?$<_nnnqe|7OlJzD^x101X=K^_aZLJEk}O1=!yB+ zi_Q~6`7=E+M70gFeT~Jof8vx0N6(4S1*4moZbwoj-txn0lQ66(pkQ)T&xG8kW0Lzb zM-i18RWn9`lls*Hrjkml<>+hAmj(+?Dc>m-^lr3#v{@g60qH7m@8K)PhgRm6ngTK7 zo1R0_jV|o+oE46lkZKNyi=-Sr8lmSD<=g#ZNX^)o-!&!Mu*kmve|iaC;&n5(aupc@ zB`*wEC!%cv#+8j&SdG(*L9=S1Fme_6rZn6+ ziv2KkTd^_NHsQ3rCJ7$n7Q|i!cwh_g>acrEe2o#M6;6ZsGU$jvk~||!AIuQ zl!cRAq%F0-cQLBafAjR;NK>W~VrPB9JD~m5d?e9z+eZl=sa6^_ zWiu=`&{xw7<8y1j$RL65?{RfhjOrophQN&rF;Dd}R-rgBC2%>7+N`?tqOBE`NDizl z=<{gEo|-BqmtPd*iux|&TXY3N%%hAs$tR-Sx2bNb6YX&)e`?0Sl_yK+DVXPmVTgXn z)9Vc8<|$_+fUPC8ys`uUHye_M7dzC+EW4{^(LJj1T=rFJ;BH~nb; zwkmwwpLh%|W+Rjj3qAHl&@|V>qUn8hNEHa*z=Ej$gp(92$LVq%nx}(|GJU%@e52ag z9-ztf8TNRFe|c_Dz4__Z<~%c3x^Ic1y|=ebIPjR5@@2U)$d6^ClA6rGw*f$6<65W|*;AJ!dn@INhNfi!#!OPxFP*gt}&z2t9E#?v{EbOm8 z*|ce7e?ZqF?LUCgn9(nqVci_^<8@-aa7K=CxFKJ8T$+X|g)GpY3QmzI$v~YPn-bmx z<8-=XwkI~OmVA}h#ys_9iV`I$#!#mU>Xp0`O*YZlypRr1#r5;|cutY?B-2vN831vW zhtAz>)Xh9v6MrK2y>pfrGi=7 z0|85v%ShfAJvmQDtq3Z4W4@*RP3wW`KWJ2CzY-ymXI)jEslv4kf2U$=(5(g<*T|}XF*7?r*=`e^Pm4*WM3KGGcyHNUR?SWT}~8!VZZ-; zf7eG>sSJ}tg*)Cvtc0Ui@%H#@-qpGv?=+=sa#MEP#rN|4$+bo31#0#cYgLgqoxD_- z&@>|brTP#u9}@1E+R^+TjVtzsK5?}F;3;6gZ0rSS#+o&KMn`T#*ZEne<4RAPIs(b0 z(suVD;ws|V*pA2fgqfzVHB~NS;1~-Be})O~8*jQ|Lm6$48utW6Gl5zy`cBVcz0V|r z@g>|;_q^#z)}S7PN4m13l<+#2zM}op3(@I;BQfL>+dSIQ#2JBXfgnoi9l#d`lVRx@ zq^*#J7s!tf-b(NqPmmo15noe$P`Zf_u+_SgKghF6i%p(6L$|6NWu_8VuEobPe;kEN zjK{Aj|1jPn0(<+Oami7mi0Xy2L&zIxkTc2Y*~iWD*h5)A(mrn`gB1ricf0o_yGKYa z=*-7H-Nuhy#m!~AQDh7HHQkC2XQ9@7j@Ztxj-_cg758T3U?_~Jo@MQ0#JZt2A-oe_ zQE6r^HP!8r)GG!50`Ve#T~GFfe_ELJ!PxV%mkSVW1ce5-qfL5v>el;uM`iB+!C#1NCk=qUTC3e?*qb_TAuEBg%AF|FIwh+@C*_yjaT$ zyuGdJfpQl?+hPf^;TL&h_P1FS`~xT7#(n*4LPQUn+gm*jFg(qb?M_bBe|Ilfvz)}< zMd2P6!&W=u_KGU-taRoVX8f80m9jmCzR3#3Rrld0lw#qt03)M-8l7j|QzWk6A_^kZg!X*!8wX*Hv99T1-)l^2&MQ8Kde`)0;NPOi!>UnF4 zdRh4#40$#sUNKm@w0ooOhMFewCyY+mQt3DXCI0hv(Zl$(R2S)lk!>S6_D>F@H(x)m z(M=@XY)wJwdVi@TEofCJ#%F!Kyyzo>G+&cW8NFh+Y8GMni0quo6~t36!#Y}q-2t`R zp8Zu7G+2VMi>0lLo109;$IME65uK!p0+m{@=-!(DRhd+lRumtd^`cBrt~vyK-kq ziWz+h#!p(&I&3i;L(l##&#gQ)X7|Xe85bFP^VX;PZ?#Wy@5npv-MN zg3iLO@aKJCfvN(p!eN?{ycH%P@`-z^rwES$&$}iiWPdG(_BpGFVGRs=f?HkzRV;;v zAl|Mgq4&NOnBo7ih~`(lrpeFBT__WIx1Q=}SXR?OddN>B+eya2?TuJ#t#(qNVu@yQRub)iaN$YU;5HWR*Kx`Qbe$I|y<^rvupH`~JLY?R+jO5hf0!$NA-)9vN^7Pn0_@9$Se>u1OS zVj*g^*zr%-U6R*QB07b<241*ULVTlI&%t11TnvAH2NOPU*ZC9)pX2F^+e-j{9chp> ze?^_SxISGs!ztGn9VluMBb?oRa{!G~BSzpQL54k6<8LllbUIR(@yjKDbD8#}h;L@Y z8_FD`IM+|zqT?hx?zqPzdaAXPgH17fIt^`#|uK`{>1 zS*F7SaEeOeQSFbQ)(MbXZoPFM>=(Dde}Fqar#?4)3X6!JflIyH_tr6$i$W3r31A~e}o0hdLU8|3JJOr zdi~bvA)a+7H`7F9nL`OOg`CA`R)e+p=U-3UKOjug1hr_wD0`Ig!8lo7`3_Sme{-<= z1n49QJaqAnx)Ki*U&Iul@crgw&~k!zF%3GMGME*wk0y&)lVQqn4OBcVT}IS&-Ahv$ zC(zmZO?gz3lmYNJke~KS2o8+FJ;}u^gINd*-4h2%>7~qlHZf1{0t=TuDUYogwzBRc zm86{DHZLSYtTMQ<9KJ!1q+TtBe}Cyu&_O)#NvE7jIan)DJT!3IiKut1&3b;TIOOrW zo9rQfl=y+#5Q#2nBa^jG(~A~y)CK|jlg)+NK~B#o@I~%td()SlD)fbsY{S)^ZFi-& z^HRFVaJp+-iCBfs@1O%`i*rm!E>l8>1t$luO~o4UhR_;n9o#RS5Q*Z)e+v0X&!RrS zIUBltVEE5o^~tXb0!@1MhcBhjaeCOE-alz^BkK}6)LgQ#WtmnGvnO@~kRxhu4q#@< zD(VQvc44<_fBcF&o<*MKXgNQsJeG@EFNYaG{b`?~Q#)imZD&vVDw>5?{JY^i0~1le zYi7!yd7F|a9&7|Hs}?3gf4P6{(t!~aQGh9_?{d+`g$;6#5P@NgIIkBXk$SS}Jlq^!QT7Re$;axc&*N4pxb6-0r28S(PwNkm zzyHT~TU6H;BVmVyR{Vf%wWP`CIO4XvE1GNs8!7*8WE0q_Hf zuc<}eW=jL``(C1re~+d3UN&DedMwTADxfMlp=FNt#ti+<;!2H z0~e#wVa^yd7i$}0Ywsl*wW<>LgAzd=gfZ&gHbYQOZd?h8e>n2_F?<3UqVlutqJAUj zw$b7uM)cLtKW;9S!_R`5w~y`S$lIw455O9PYzJHEj?O#50*@$%=MaBw{E*RyUU_bVLCX*=zUe2bMYnnN$slzyFC+vg5;XhZ6WKNiFxat&nKx4&%Mdt ziN+EizoyNuf7l*B`dQ;R$S`A@XoaF(6v-+ z)R>Zn6Obqi*3EAmV9rxjU3^f}trNqgu2G!i3nv|mecvUf zBhi@tf2XO?Sy8^@2JLj+cP<8iiSftY!wI?aWzGC(p4m zXwugGE}_K{tK5^`7E2w96BCOJHB;wg!wiD`Nw&M`KBYHo8HDyE?~8Ytj9~-QwQB^1 zf5M}wABCM$a3)N&c4OO~*tTukw)w_*lZow#t%+^hb|$uMp6}GD`fvZc?yA1(ebH4< z@4Z&mg0Z!3`p(h)=3F1A+B65N#bk2#s7?lrzbP=N=aOTKt zPLk)h@7rGnj6-h?y^e>cajh}+ZVe=xdBSL0up&XIvxRWOy(vYyOVX;BvaE0LX|eFo!& z*|G~4QfI2y2<@3~CWQ-;1|zy7|z|Clagj2|aFO)Pvz9~TNyt4Ue zD!0QOKvXAA8*?R?Hy!rW+Xj5@+`}~*Efd7v<$byk7kq8cHw$FA!C*JK5j(X0-CGE_ z4|UfBhkwxp71#o(REpuMJhkQIB>r)Cgu&8e%?Reo$R~p|E$epiD`AODJ`|60_*2wn zlJ##J(KE+$QsK`E&!_!(+&F&?FUv`n9tq`xOUp=vuUT++&B(d5WvRt(=*Xv9<0DpN$zj<&d#Q6hk_bl# zIKM#nlOio_z)8_P32PV6~Xiiz4PqvDj%zXwg<^qE7q zvvwCwOJ%&)2W;YxJRK*@Zq@@U=bJ-2XM9(X3t3X-Y_NfE+EJb;?`y!lUP?5I^GlprSU!%lnrt%dDU zauD+l=*M97PMm)pH{J!(GxbS4PGzOA{?de9X7)I56_As4 zx0u78i-)O=RMXLtBj23v3(fpQRqjvQ2iO?9%(fb&$bD?ZC58Fdp|a zUN!sS74!!cNA_WCSTER)(sZR#e`3cURWW!|>p2+7kmzaZ3s!2o#ezExnjq>)xQLu!%C*$LM3HKgW`W7(7X&^*%WtS;xTR{Yl%r$CFb}P`{?)>VF5H zYWqt(wY{Z3_l(^Fh=MR3TZ&&ige9G4&8e-8a^aEx1 zp99)0F?NwZwZZ(>pCIVeO3LQ>huX!oJT@pmO+gx~G)zwcm38#itJcTVRB`b{yX*0n z)8gppzkFslA#rs$LX!}dF#IXv7sa77u8XDJHs_tt4%>w*Y0%*r?r$hXO!d zo^svr`Pv0v-5g&|a3$nLaN?!LS1HR$snG~d8Oy7{Lf!32zKwbFNNZO?lk~VxiomyQ zSo1E_*jxbURQc3!?SVeJK{uL&k7p(jN><9B&}de)Qjmma|8tHjAM~uNc;Xemcs-%o z97(A7kPC6Am$x!&;Mt%O_Vm)BwESK|_^9>n9V~;laqAZ=85>4SCL)+oDtb9i|KFhu zvaxrz+P?yqM2o>3&3Bx>0sifRk+R$9{1|zIE<58HH?t-4`c^3+VbvQFHQZZ3nBJL- zHx5y5gp0pj9fp%v;bWC)-@Ac19R6bC!!g`K@|%OJO)zEO%gd9&+8~%uCuBG~0_@S& zI!1-7L*fb8BQ+CR3ipP+HYJ5@%fm`nq6;x`OE|{2N~%taXhNN7AWe#(W8RH{Y26-cxQcgO zb+ZAar&0xD#LdM4=5`>Uv;iP~wksyfremK@uvw4pPl%_`2|dWMW6pcntAV~PH=UwQ z@jYkg7SD;Zd0XA)FfIw1lCHWf|GeelSuVeFDO)k;>X#2;WR#YW1ClpK6%`-RYCafL zXHs92E#gLT)B98@zk8a2L(;T(v#;d}cJpwoikb;YB~^_YRDoMRMd^LPS=6|>1Rb6y z>c^3dy}z3+da<}TJPA^8W#l;3%;;B1-C;*jjKK#^;C+H1--(rHeLl=Qm7px+f?v`9 z#v8zkcY*cV!b|CA?2zV?@f#Dflgt&O%kSdS>0er=wCp7OtLeQ2{utt`aRyNLmvr*& zr&LIgTbpa^4JBvVhfmdPbzjmNNJN!lqSL+_romh4PHhtVB4u#^ku;0<1C!whu%1gS zy~ea77svoyiY};~L#J@`i_5mqC7e1)B{6O)9icximxOsue=z|SoEs-%TR|7 zhwP>i>l!yCjZ#s6-vO>(u8=U#hH{L%BxgjL8HMjwlX0HV~TZJ=BA-|+8d2d}ZCV8_< zcKOBPjBtQ{5Joxzmw9TZyQm#~Wq{385DmsK)fT4O2rgrT+-{eMu~=nV-x|DcZY9Ah z!7%@WaP9%*RH423mJtI*p6iSl_4vYjIH_Wc7d#;d|GjkR`~8bs>)_4qcy-As_km91 z$?8nXRKs!88gjXS^Uu=7xEJ_D3tgM{e-|Pxfop zw-l1A?6XKfVtdNgzM>aZo7++hg5+js$%0%q+lr6i>ozs&FZ$eIYk_j=(us+sC&!fw zfOwZj8`oa2(G$ zt$sK#p!-3-i)$(k%pt{yn24*rPdtRXW8_668ub>+aQ{bVnP!aBo*3T&zDRF-)na$O zutZMfSn%A;4+*s&wZ!xQl`+jFB-%U1bn(LwNyr7YZA=#d$ z8i!8cQTso56fFUfO(mMJqyTK8?@0vwxkn$7l-YL;&p1J$>uC$yhuZf)yw5HH{EYa!yOPJPAc=<^r^Oxwo^6#~ZiDL4G zYW8)YBQDY5lztiPB)TM9o$6LpFv9ou58%2t))9@Ei5uOaO6d{FXm5p~r3vtHeGP=1L5? zPc};9gqY9*;z1>=;lU}6bFNt1CHei$k{6P;xQF28RDwVtMuKiKG7NkyQfLI|?17=a zaz-RI=etxj?9nk8z0w^l29m)d*b*q3&{IYp$q}ttv_=z)Kv5jJvMeYyAt6rAW%N<< zLE>(9ACKac@H3#g1gV=Z2_-1IjR1or24^80*E8X1sIC#0eyYs)q0zx%$7zppaW4w` z%Y817mEWeo!04#i)feTMea9IHMTzjkOU=r0SsC=ulGg*CBp^(#S1dxO!W113nzjW$ z&c!*WT6!7n_|svqRsOpF#5y4j?S;^}Y36gywKgyGZA~0Qf0=pTWbf@l?i7?hxCyIa z&?qb3%;sI%liEqro@SNZrqT$axBe2$HugvMV7B|NDV@>ytD?6=vhyD38A~Jh5e&Ok zi?B>CEZp~dT5N35VI}0+YoEb&7J+Hv2W1e0+0VRYc2CxbRNz z!liQdEIr?ySvT+)4I?tNDj3eCycp?voRr4E1g|m#-HdNS>dh>pWe{~Yrt+=1_slli z3)-e6FrmI!&WSu#suCPHJA#enFJ~a2#~bh)>d^|GX|bs;A-|oRm7hmAg@9-$ku; z*-BBNa<~oF;c6tykE6d?-E~PGqoZ|F>vxPqQAv!j0cgWJGfm&(z=SH85{)7hTa&zY zb$zmij_oM#Gwpxuv3=cw0ylX5>*`Mzjc+Ay*wBO_3C~Jh;%ot+cG*IuYdf{q#znrv z59jc%%V&x=`@LF8sEACCH99oazw>}B!To#5#jrNIpbo_f7~4yCD=F|__Qw+}?VB^s zIzdGh;O6Yi0FBUonNTs?`3 z+=lB{6Kau76>x!*vyY8BN!D@HinM@j|B)4fix$>OcUpLP9gL%63-xbnK_(a~S+|ti zs;X2G`eB^+*wcpk8wW7MJSpBevEcNi=U%ro#xa0(Yk4^^X`Z-E>1(O$G0{F}f;ghm zSA#Qe9}3^BJ0uv}3O7rv4fEvL?nyPy-)m+AjI}+x9yvBF&hsyI>&S5Lh*x(Mkrz@n z8v{K>Yc!Y^h_5n23{sFYN-mU$o-?U)gsA`vP-ZmYsMrf^eQ#}gI<;RVSLI7AkC*)qFn6CU zuKC-c{rbW*;U(?pe8Z55Aib}foYINv8ROjw;Jnxu(CEsHzw+@B9pWwBoNsRxi`^#a z=plQI1*d9{9Eve-JrtRNoteI!dnD3$gLH3ZImElzPv*CmtyB|hG!u408vCt@U>fG# z?#2)Nn*rW#QYs^Q{ts?8eKR#*bqJutUyPV8nvnh@J;9S61vwh9Q;r-97 z*IgLFYq$)9+k~iEPa-IQd6%N8)g&sav;Dgz-dY`je})}%WsY;90^_phspfXNRJrqq z-mv}oEerME-l;c{RwYXJu%FEQJ=i6VPY^FK=PiZPq-PgC!w?sP(A1xSN0(~l(@*Jx z32AejyT;H3N)+vk?oXG%Rj0QZC7*3@LB@dUi`Hy;MK_%#^Ph4FuR7?Q59~~kqXIHy zK39tB-^J6%ATWwk5$qt1P-}AW783^%XM++MiP(&-#)DMMdK#}ocJ@%sC~ zjCo>w;=xe=qs4a#Z`&>bZ&{A6QJWC1h9M<72@&M|Q+dOt@(leg(0CMvk>H8vSShc@ zmF|IziJig*)D~t3Z}tW)>zoWec9jti){nFGJSNGRJ#A@wc0j}8x)^{x6d!&u<+w=V zW!9y^U(B-sWbN^n%ppMdRjk*8j#v(u6ZQIXuh{_Uks~o@8KePaj|z2r;PZS$3|`dVt?Z$_KcNn=%fe z(W@Vep5dDo+E0tIMVH$RExYSKXthe{QqcHSZd&|n@v$SLE11}1++k1k_o*Bdhqo)kL3pWvvmiK9s#-jwVxvnUH@)L45! z_TqEcSJXZGxI16%s#s4^ks%{u)@u6_SNMd79ln0)a_P^aEjd$eB9{&P1tMs0GpvRC|C_L=fVDO1tQ zvKU}6)3RwY;W#IBJ!cL+-wybfg%EsB!Eq^dTD3|J_ItHsD0e}ivb73G zU$FjO+ySlfzqD((4iZa|4I=^?7~q>-l@O`I%TjD47}ytY;E#f#K-E0#3Xs zl*ayWhi-j=b3wu;rR{3FoF#?|`AL~DJx#JcVLg&%zM53(c9@CClvZYGOY2>Nl=I3> zVPPm=1cUL1D6P~X5*m93a&{Ya&Rxbrk_OO9CfsJLWMyv4&Ta+Ke=#{G(N2YfmZmI~qBh%WLdi!N% z(8ZTI0Q7HLiA`V;E~Abxd55H89hG1L00RqBxAjh0Ef9Y!`9Xgdad@YT#l^sJHslOC zmBsM1aCsf9jU?TrW4_&?w;NR z&L)!GR$O-xnmv>$*4dr(l*LUb$~Igee>bOrKR59|Mk)}|3zp_?sb8)j$|f}g+*RQ< z1+{aMMP_zdZ?r9jX}u7!z*6!0H0)zgFUhEU1hSF(!=_}&2G?1^ugZ?K<|fPIs_9eqmwq!RHQlzijpCMZ_(-WFnNBc`8#y!;8gqS*2Bz zK@VL=YEbjqIC;s<7I~tm54OyrMBa@()rzK$jw}%0Q-fKN!2ULReF5#S&;ts>u3wmv z`DfBc^q@bLs`E-~jPjnG{4;NX@(8F|Q@L0UO<+QG#RGK7j63XRd&s(qwZbNKise*m z^TDHjM*t5y+0cPxj;%iU&?X#sto(G)qsEdH1`!0z>Gb_uXUCQkO9>9a1k#CkHgkF0 zzZf8OH2=Md*j^>%6~RU`bggUbU^r_YM9XgzV<0|M9F*+iEldZMZ}dh7P6jzaFJ@9p ztAcZ}O*m1i#?Gl2LPHVlOPpWYkZM1Ttx$$JHG?!fId&LBCN4462>1y=vAukaTg=_p zFJO;pPA}0_?c^iF0q-serq&W&HmA>Yqn}Nd-;( zYzU*Pp7*CxOy$5?WfxUq7a0N@(sD}34o6ZF%60 zdi#gWs-)qxvWRnpro$M0(_UEad|?NKK(oWvZx-UHI+Xpt~y20rZr8M^0aRy{;R$$d(7mj3`6fCj`RslV6Y;5Uqyi*nluekri8k6?U;rhux z^>3SkPGP2E)aVsN$ceA!`#(pgCY;G2CfK*n!i^Xq3?KgS+3iz zwT|4~0MahY8q{b}xW`?Xse>#?XOd>P|2jyq`wd1_>W&-6%Y=N29$8f^&B5n%2iTH(ziC z5a$}uV6_e9QQ!c@w$4|vnfovB4)^Ct!8k4 zT!R82Uu^_SX%*L6ucG^kP64ni(j&Ie9k)=>JHnJ>tmUO^ zwKWW$&H$P80^MP<-WZW6;H^eUmg-8Y!KznylY- zfxcZons1$RS9O6UX%!QIcbn1#YS^0Ucr|O#4RMor^PUj}E%Qb*q?*8fg~o1sKj@dt zFF24gG16>GClv2xMi^UWn+|rx4u_FH!QAkX5|i;;w> zO|4LhSK2@cJC!KXjl~@naZxVcPBtjc=)3N=!qPSiR(c6$?@|dgw_>Kin@>N`RXZAS zmgWZm%Aw&wb+Pz1mCc}wxKT!0SPowCVo~xid;n#S{!qp3nQpi1P_tkR3yM&5D_Dc; zu;c7QWJZ?`t6rla;fDSe%i1kA$WsX}DTE4#z8N7GSBPxiTeaG?Izs?Ez*ZlB8@B$3~Y5hWK6x1jly9XgbV6Nd|es}6xUm>t)LJ66uG4jyAx^8RLpr4{4 z{|~ic4e;mv?9qyQo+902%VMx8K=Y2T26rRo?lQb>Akp(g+D?7?bV~TO{cZnj+bKaq zxgud;z9^sE86-VVoaYiMxg`$$TqfFFRN3Z>XY>1W@|lrD`> z7MGCp?t`gGC5OXWnX7>S?yd9@ zj_AN&fei8DP~AC3f~`S@dx!@{BO=c6(AVIif?+MAT|<&V{b#c=HO}U|MeyS3+W^T# zqQnkg1$3946fzae+)|kP1_NlrArQ|kz7iwLpo$eEV}4e)9noSTw7ztyItC|uw_sIp$$ zp;zQ|NEQrba2f}al)X;~)biU1)EHqV3_3Ukehz}K1xt)G^`dk@)7I2MR~UowYBq!T zm}-E^xPUOTH2)s*$HXcgJy5x6%1(+8e3&6BusUj`c}p>7AP@V;WBcfqkX)2bQls49 zrj8eC8spwHsoqN6bV8zj%2^;Pn4>J54_u`iJuLh8eWddiknD*8yl{R4BaWNA20#eR zWfN7-JD2e>j4Sa0q07LZ`}$L@sorK|eq!r;``e0CehNse4z8~p)}bu{(CGPBul6tg zRyU5!ZKp`#5tHlg93;0)yN^PLT?9%X?1_XmbGlQdr+-6e-@sYO!Yn z+fHFXfC4$&-#Mx~K9EfYTHP{M>^<)X!_nD;c2l<0ecxUH&rT5~SXM{wb_t&69`G4+ z-X5{#z}%esp7T_y54{IG1SG?B?_j&rL%ht`%M6z|uNfC0Y8c^W;74M4+_g({=^(N4 z^%XR`;fzh!r_r*PrZnA>Y^)RVTW`hPFgPAiQvq6dt*L{CN~8Ca|I8qbYist!mBTki zS?&$(-BR_I)_JMB!p8_i?_YgZHi?F7GG-Qv$c{)2n!?e=YHbc&_|EH|HGOWA{0HPa z9s?(an`SNWYQk=(OLvb1EBcqq-b6E|ZpKNR2AeD%#LTJVP{jr|eVGsQK^u! z?P}n4@#DLwQe2$k%g(o7*tg?@Ibe|xN&)OqpcAAER*;P`GKaYsPpS0Z>K4l-4A-d% zmtbG1N$|JkQ$Iy89QE3`&DhjKXBzQRsJqVeKI1;(omKbSBnLZD6x4J@Ra?=tp% z`8En9eQ$EFSQasXannFvAS!~|iM3Ehl^A%Z@KUNj6Ub2O4C1LKT`y-wao)tc)s03TUmMW8j)Xzd?LD=;Q_62b94*W6at#Ncz;ZwZ6VxVF%IGU zLLkQ4KeNv|(FcByQGG_WQG(qh_0d$k+Ei$C8n``2LO8F-TZ+zG;`w+H9r$9%K@Fs+ zzr>9Z>8p_+7;y?^nL^gt!C|ZzKgk*FE}CsC=8DL2^$|G#!LHuD^!jLB{h+JvtI*I4 zd{58`nywP|I?4U<-g?H};svgR?*dl0QyYBU(MVoLt85PRagjMnvWOD~2qr#0S*& ziJ)L>rUbacFVJM=WQf(HZG-aF=860(b%9XcoI7Qt4hiIEEJLklpZ!v#PY@V|q#e$p z%sa^o2WB?k@99#ghC=0B;a(WbREFS1K8o9-9W4j<4^%GbZe(`92g%KZ5b;nHP7~*^ zNnW6W?0kJ}5?ilu7fk{0Yn)0yLJ10bok5?7ToA)bJWV!JMtJClQ#nsnafnl5NAN%3 zT0p{Yn-Kawa11QlOgp9K{ZpWpSUH7MMJT9t;jLTq?`LJ?ZXwk$_j}g91q2r)3+wth zcp@3Hyz2V7ePQcYG%c@f*=#zBIaEp6@GD)CSVilU%)~iW?G#-cbggH)p$;({+uht4 zWm0`3SIt|=;Jb=MHnW~><5VMU9Sk|cG-B<4IOrLK8zM{wrtxB`1L;6NKb)dA(a@t% zCJRK0N2uR^ghSccywINeTS6fSfQv@r(2~1vn2)(DGdVbv=x;wEEx|_W#%Di9B*`T2A%D z|G!W2z@G&2b~82^)|Y&8)B!b$jJ6=VV)#3H@syBy#53|pav53RU(#~`_0w`vwCW5; zzH&v4neTpN88)YoOj2|UTYHGzW}3aNt->^Uq!y)1N+HPB*AMVQifOm_E@;n&0!nU` z8XhL24)lw*+jNs;S$Om@32S`#e-Cz_=$*xVNax`=AV^7o(U zTRx1JNK7C1KJPa_)@QE@nhi=i{j2*e^TO;C9i2QDJ2?_*b*$N~d4kAP%2N?)dOvxC zL+mv_ZUIF|GY;^y;SV#V)9gu@hY7?nK|^2mP|+VY@)~-UD6{w(THB;%Nte&kxtTe!DVyMlN5=7;N^6NiHpx=0gQuhPp0ru=Vvyl@s zELNQB0|)*R$Hg=N1V#GNgG0A5vA~E$Sv~Ac7eiwU9R}u!n{zQe;i_oT23mVEfLmjL z<3F(IIDJ0C&txgo`1l0G^-s={uz~X z`{CI~v8q#B>FX&wy{NQ@zwG`ClE|7FxJo+fer})M?1hELJS&I%W+b3$wRn*Tn{$kx zO`xA68i6Zb8K_b|Yv~|}80|oB7^%&%7&Awth5!`k1W?92Ysk?a>&Uz*-zqMD*0SY{ zJ%rcz@CV=}Gft`FuEwS~DqigJ*m&mVuj+(CR$;TqC|<$B=GmL+hDS&^IMAsm z62zgZ1l=0a8F*0WtUTz-Vw=MuuE~cFDyHXG>%P&)r^Fb_5emL zSb-AbC1#eH4ip`S5us|0EyaNPY7a7QB%#lV7f?wlk@9z0zB&mly|Z(=H0=#Mit$z^ z@u_1oSBEIPI(Y$uN4q=R6v6l+@3Gukq0*RI2fVlbA}S%H5N>7mXf%+1bliz1``E0ZiUrmm4p4=AW)ei)n$ilQ+Xxz3LM z6R}%?Jo_8@t|A$C3=ofRtk}}IqthpLCOP~Dks0|)PND|=>6A!3Ol)oQ1+^D_c0jjT z0&ZP(P`!jutkvh8YHTW{yPfG`5sptl!!-y2hzhYLy&EGc4cU2}KmI00D38d~ z^zQt1dT)L9pH$~fG=Qk!UMyN+Ez>aytT#g)5)K8K{lF#^?ac&oGL9C3NQj=lqK=j? zErDMsEq!xq^A)|@=FK6KkRC+q0jPC`ocQSTaV3O|99Ucg-k3nXyG%(*puZd5lsOH< z8mb#WKT*$CW|Legp1LSkUU8ZR#b8OO8|N7yz`y>|M31h78fo%(Wma?w4z(N1)e--& z&3(+fwe!XGgkX$^MeeUjBA(&cuYyudE`pX2tR3me-s=u z7gIPu)xT)*!bNA{Y!&rOi-+iF$!%nX^ylSar*)h3&5`A8VQIj8R#Y(e7bD3JJ6~S! zBh_PYxZ>CC*mrJ#ovU!wU{E#kXlHLOG4!9mjaAFIUU&2a+BH*5XINXMe1Np;d4J_A z@OP#6jl8Iy0;e+;|C3qCS>S2yWA?Y?&LZ-o=dMnxCl^DI8}@%F#M{;Dh8}I4N=<&Z7z3I{j;j`}%W#ql=(z>;)91N5V9+-}Ra%5u(`xnOO zjO#t;+b&3>>Zo>Buq)u<8w0TaI0*OagbYRN+u5}gBcDX+!Ns=S=u|O*FqsJ@y5?+{ zu+CogW2>^#RoH4Y?S@><8^@h>?K72wT&C`kw!8DbJ z-0W3=H?DZ+3vP|M97wqx??ut0Th-D$LRs>79){#ms;J;tgQ>qw)aC2l6TQi=H=yuX zv(VcK_S_zmmd)iA8Q)mT81jm`dQTf*J%6~z!;S7tv3M#qu@Rk8=|5(1j}(6kkH}cd ze$;8wOQ7UaSz zszlO?Uta#O0E&dyVgeN7J+||D;!JsPx4LjK!?RvZX3@rRperi8yl&Le!^j-N@=14~ zVKwPBmPep<<+?^4cOPtFOD@S*uOZ{i>YX*c;yetMPX?p!154>w;De!0oD3aW&Fcqb zm9bv+28rPdSaq~l6G9!iT|d2^{_6|LY!=(UcrDRy0ES84;v+(Bxghw%7w(*|r9F09 z9=<^OT&?{3{C>awY=!I+A4t1HcXOQlL^jm>WLF5-H~V8g=8;rzF?Y%uR{<$A5C>-0qE{=!zu{8*d7M>n zi0RK(61fHD>56Y5mi0fV>5LhnG|R@>5{G)hTi{A`4%TfU;wGoiG8Wsw+k0J>lSyX& zUF$l-+Jr?Ni!#OBGi*M-lki6V5y7ecrg1vknyLTVv?sE|np;I!Rv z&RZS}y9I#*?>mSNz0N7lOCAHY3Q)h&lbLGdu4WoL(34NlT_32;DrCPi%i`FPuKhC$ zFkpn(MII7&yP%=m3xq@0af<4ZEUfQ3uHh%{t!^qtt4fxdj?UrV`WTB#TxGpfvKE!B zMeBO;bVVU>OJoauhD!*eScBt@WwN@#UkviMTJWCfHPZp*H2bqCMq2(;YM-;PqMC6( zAF$izZRNLS5LbDE)!O>>dYd76NqQf1aQKYVFqXf7B=ec|TaYcWHJk%WcvvxST<*w# z&;QG@0#`W=RarP$S$H_vcqBw5#aWrTdBnsxBqW&GxY#&YM3^O6B?XE3|G!5J;F#ns z9IXJ>#B7|b#Q&FmQOve8LAcq4C!qC*L&Qn6uK6-Y1hLfs>-#!PsRENoP+bz_P-=P?p9| ziDkQsT#v-|o%^rusLH?%b(Xrt$S3NsLSMLOA+M7oF0_;Y-Li`Zqh8%DKbOliDaT0# zlfW+;gpu3pa<7~3jRk($91Y5D+UT4r23UJXq+uk1L9}W?_Wvg|!6IYF$6kwM&7mlU z=|8Z_DKbxwA!g?wg#pgO%Wsz(+R4ydRtDt27ssR|!>Ph1cT{N!m6I+PLydS=hWn^O ztNH(J3}R0LQZC4cNyFPvI27jq5){#h4T`fMOa_GBdr+nZeqPp@v-osP(!YLw?2Km| z7@&`vMJ7YdsG5Ct>6fja26%iH&x0RaOEH5U-19L*^}h3$kv%E$IY2yr#O)I3lPKC& zRf0ay5{!JVM3mty9L?PTt`;T^|GRQBwS{v7m|(gByj(3T;n+BM{;MPx90i4hk|f;! E06FvM>i_@% delta 63485 zcmV)AK*Yb8uiF9av)ciXBLgupIg@b^D1Yr6%XZtib)T=$ZlrT84hg=yieoz+r?!*Y z%4vHhv!G?#=7b_0l8WQ``ra2HfTE_E&Z3(x6q3LNaNjRD{_J7q&;I(#{}0bMt5-k1 zje}W`dU>d_+3J241iqIf@hpuKFV4f+>TdS)e6>jYd0Cg6awz_?ScXwFKW-LFm49aQ z_Ni=^eY@=z%Ac=G9(*ou`Q^SVHsvCS=g&CZcAsUKzZSo&-p!U_;Q47fTL!t8tC$Zg zhsUPfAk?<6)@|KBRO_PV#l`lnYUPzrT?GF8vHDu}Jh~~G0*4+JOZXR%ZUzBw?&!!M z^1aX;nedUtc4#-nP^~pAoCk=%^nX>4i++3?rnAh;lO!TGCB7Hr0ga@*!6)Aj88TMD zMlX-y37|`rC@%$8F5}z_QY9>Xgwq7!w^d!Y&BQ5b?D zHIWizE{-pcguSt+LItzF7bOADe*AEG!bg8>s;`U8pL;CT4~xi0t}cP{N`J8P6WS>C z;y4*?b9v1mCs)@u{FVl}Kg|Sq&r?~?%Lih?>nC$w-!|SF#7;7bLodwZSrm9#Bx?As z?LUQa@bp+UqG&hOrfr&bnBa;Ey)A*ei4F3H+@MrKG}+a;k8foeynUCIl6!G9-E z*zcaLJO~mmj5E^cR5MJ&mCF&1g!g52zirAXeTT#Re1tAaYI%9}(|^^HxUb@Q+4a?c zOyliFQXr%5jj&EI)p@k%&Dq`5?%IsIJ`(a=@KDuL|)~ zY!@LMFjAy2@Ds(rE9BPuj~DMh-kb@ZXQyZXm{UNkuOJ_fxd(4F)GI!O1)f4>w=a;B9Vozw!i60bWT2~(Nwrd zyde9(Bl<3rKo=2=kSK3~y(RGW^~0v9>YXVhv6ljiEr)Vipgn)B?gY_x`x#u=luS~} zgk4^LPuyPKY-RHw83{$Mzm$D@Tz~n=SK%Vc5&#wOhz^vM^$+m=6K=b~ zpM_b+PQ-VV1Sd8K-M|k@X_ck(0bT_cm1R5)L3cjzH`1_qLXi>n9fB9MgUU4w6zeou z*SOD`786hSpq`NOpSgXyEA!z3pn z8h?C+3?5JE5R!>1!H>sbHviWGnT&XWeuim6XUTGlMfgdiZ#CApybbP(3w=!|LK0F) zMUC_8L3F}m3}M%h{|iMs#>fYrQw=66Rm=|F-WSUiBe&HK6) zW?@>YcOw<5=wRdrEcvn}4#3D*xVUv)$$$4}()YV5Y6ivxRoLQa3{LBb6&6qHzAoa7 zjvs7R0TMCrcC=2LTfQ^5YNy?Jkj1aPb8!}y!huY0Rf~$<>W^wbfENa4x$Ya4*^^`x zt4Z#|@J4ngDRfmK3f}YVBh3;z$iUrBA_XJ@R25ypKX3&62Y>|T^s49Ys^RCRGJlk_ zO2^T>*JJB)NJYgO#3e`_WewOy8g0(_QkI=S?o$tg zzqZWze zoXNz3wfvwq<3ZsUzG_8&V3UAFDJ`}9jDpkTKT5rVzA!ON$Tz@HEQ2z?x zo{5Uv1EJN<20|d>zN3I!5;27KXPihSf(!84_6x|rI5EtJ*FEO5JxDbuw#4SxX&OK@|!Dr&@&{;M-tGV*T4uw1~ zYQY^(5AALfogL^YRADk(IVE>2;ir{h?JY+PpgqHin0jQSbLp2JzH=xl7RqqMHFRI>DXZM-mNGq z!inAC>y#J4LpvI)%76Y4=NpDwAr$EeIW_G0;TtjQe&^mb4beW6vyjtM zhFBP>s%@7y^@655~G_-r=E?x1Z(Qh%YUC#HP5mb2-_r^Xp*pv#LJL8dT=?3k9Y*m^Y?xG?!hy1rW0Pe5fmA?O zD>C>NQPko-Ih*V#?K(>i<5d!RKA;2{qNj3+)}40E1Likz6l+J1uOD7m3W^d{9I*>k z_>U&xB{bU$mrl~1|{G8x#YhQ8x{pcIk!wTTi= zhvkU#Y=5KwdmyxfQ!`0_iA;x_rsYa;c3p@!F||0VrDkWPv7#LLLHsd0i^&-GtXHDR z#u)JFfjDf~nejkLrLE+ocUHKw>c0qTfS=%Xg(=~M-fyqm$6-l3O%~fhz)W+0zM_sA zhbztJL%kmzzofHRbI_viCsAPzOLu~U_CNq&+J72dWmPi$ZZ~oSI#)G*bG-v-tmJN( z@~ual!1U#V4t*wLES85lK$VH!PB$w8FF)!nI4wZCQmPr;&UNb-y{zZS;nAcrbd+=$ zGR(4b4)&>CfFR6)zCb$eXE8B4Z|4{c(M-`My}W}+G?@kBynde0$*w`((~R0>&ldkciLz4T-*0M zQKFUb-O&gMD%5Q_tQh(O2?$b0y#aj)*xjmKGp~A?kB&;gq%XAem zMZr#Ix}6I(CQWTJr>?=z74@v^l1z^+6vl&5-+}VDAaJ{J3LZ+dQup2E5Y7!?<@1sB z?M(TyJYWW!o#AV*ALfbUdGUL&Q_VazHHC3ic5Q|W?yPb|dt%cKhYd_U3h&>0r+=_3 zKMy#lM;#os8loKrPT7o~Y-#X2J!4Cwyw5RKeG_uT2CjKMXkBV7J*SBCu-krUO)4#2 zVjR`wKE8|dS{~LWY49nokjS)3*G&_nL$z!6Da`jGZTrgYB`Q12OX*bX&_)w&xBwzx zsw5-=8XLAw;u^2uGLg|}r-<3>a(_j^A=~4G)VcNZ@sBUOKqtX-p+gNm6E7UtuP_9(4V< zhBCSCLHC%Y0bMzwszfkrUgFf^#x6|KX6~id#u{{2Wd~CTMs}ZEyNJmD-Mh8ORY!pSs%uppnKtFo@0oQ;_XbJMDUZ`9v+csV z^wS9mkI7w3ukQgx*WvwsYhAY#`{|JMY?Is+GoxHe_GBqeML6BWt#KbL7;W!#=%r- z|0-S0ANby8nJW5o4ig5o!!IR=-$fF9-Iy|3vAdl@SE6CwRZ(3O+11V6Rh~v=U0s!B6(y#=y4hX*xH?}atGnxUS{19Eecaf;fq$dx zbzYT!tMBaN30mY>b8r6|dflF}%G6lV{Dyt8MvdA?r<47I^5eCsRu}uw+rM_pr3d(D z4<+_*3)3HHsIGx_4U}|Sm+(;N^NArJxkq{~v zsb9~*-k-S-tTHvW`8u@#%e2P_J|A4h6SRqOFCVVgrd&19uB8pM55L1QcQiuFCutvV zp>9vl;{(CM5E*a|n72FIbrh@rgYDfyUC%(?V>{dNSNJD96I#|SfoFGrofQ>M&`@K4 z0QCI8XMe&?#U`{!`&hdgIuV>EFCBm_jblN{V|7082Tf>ZJmXOh8hpbC1ar0yrD2I3 zK-t2+%$^i(^@D8niMOnSGbv)6kk16F);+^@A*3WSDw5SB?0<8MlO%XD8${k64RZu& zpwD68Ejl(ke%erz2U#h!AY?w$*u4OMwq2qWV=H3*D~al}Dqh{(^Ax=uC|{U@P&7BI zn`=v{c#aK}Hi9*_qZ$K`yeG_~qr^34M2u&8{3%W{W+YG{>?U{kXQywLX{Y5)mcsxp zlG}dGFBe`=*Bl#n$URlxbwuX);?afp02cOga*QhgkZ_mEHZm5aaH)yD2z~`Xme4;M*GWec%wxV z3KWmpNf5l;dX{>XW>H1Es`9nTSB&5V1v)%OgCs&stV0jA5x|AAuKScA2-RS%?TVrCZq`2j`G%OAW^4?0UJiKP@v_0NCT4q{RsE4}#7sV`xIRR!nzm+YUJ&iBu z*6_P0rjw@i>PO*TAf~byX$@X~2H9`PK{z$keH2ntoCya>>F4f$jJUYvW%NkdX}wY{ z2dzrjA(?n!;Mv7!fi}kk_|ODMV}xSxgq{9C^O@8MFZ<|Zhp%;;bZ`h9YKzeJS42>n zY(d$wL%9RC&ki?_N^-Y@=YJFCFx3O#EQn>{7PlTFum-iy#xqc+12CDU#!tpaZYT#N^Tv zM!0=imUK)f5k1WOV0^FDmGO8$A+0hVaXT*t7uE!)Q6y7yY-(GW!ic8A9JqU;&RU!o z%9@is>pYhaElq0A8wcZ0DINt;DFOxBJd)%%_{sViwau-6t<*;cukPq}GG^&PC>8W` zpwAqdsf<^-_KayxdB#QbdT8D!%Y3&=fcHY2*dAC{njpJ63wN12L4iB-%%JIEo!N72 z1ZuT~i_8GVxFziZ1U0ld20DpT9hlQx9|Ge_?G=H$ zU>G^BCFea6l>mQPyW_Qw{RP_+`!i*Q>ahKp=iks{tW!W49!x8e4f=~o3bkZ2;NiL+ zdcn}#(bCM?yg>;@8uDc214-k0<;=#7$f!oB21RXu?ADwnWNdqtMA^BcQCUGw;|fmS z(7v9SStqg6)Au5@gk8p@2|40kWFx5EU)<3-kvdS-mheveaL2Hb^P(`3py*S$!o<0y zv1YAoR2#nfCeN;tB&ze=fCJP*YKo}HQ+9n=EKcG$Ufq2A7O>MkynXlf{o9-CI$ph^ zk`M2HUR_&@`R1GNe|RseH6}8(8zbRC{`b1bQHf1cD=;rRFeKn{dp-@#3D%vaR{uV2 z_V!@tJzL#1o8i)3CzYrK7QU*k?T#ixi=8FabWSmoVOw?_ zQTI9PX{#(tv6>EGu~>b%7i)89!0<^cXF1tN6oFfDq=BO;D^l89()v1Q{GU_{!MTTzUm$!=Av> zTU?F{Y(#_TsljJl_Y=LefE9P=PLO=Pz^R-?Q)DaksCCz3G0;&2?tSPNI5*pCr}b_( zp$iGR3fnI7@pN4yizWZFFnkEem>kT1RjYCwnk8cp+5WI$&X8T@QC*e=aE7F|4B;w` zs+glc$qf6*q9ngst45$^b0h;LPKl>vMaNy&H1wpmAudDDHa)nsRv9-h$rs&ft>L$s27?GuB#*5l#N86gAi3;clB;225B)x# z!gBRqQE2Clt`HSUaK*6wbCD6-a75LK!&xN-huVqqQO@jc$qamkzP*?^lW&fR@qo&k zfkiuSp>Wl62adUo_V(JAkF5`X2EcoW-!O*rV8Kg2Tx%1LH)-e06yv3_9SqsfvNy!X zgCy`!xzj=cTZZjOTklyph7x?iMM{Zi2UYA(z9HCyxn^B|h2uTr2@Td8@Er~S-hT!4_K@PvyWQB^Q2p^9z5g2 zNDFNNi||SHDG;$2mBns$n;m?+gEQEhQy99Hk#Q0e3b%Aoxbak7L%K{KX#73(-ONyV zFLuf)unYpH3OWzkBXd-LRzi_tLxVjcmF?AudYpYcyTgNT9!JiP!drv(S^JS2*}{MK z-VBiw3k+Ct67DuiQAkTCzTgpK+RpodY~CqVY2hOLw&^#@gmY}dmM17y)wvXZ1x{!u@r8I6CnWX} zOT^wj+BJ_C4E2UXkU3~hKI{-PucPiWE4XWfsxLe{+M~i|Goux;qYD~jqG5jxgSu>- zw>K|-d6C%1_$rY^9Fvz(QkPfT3aFxMHg5K2&Uold4U=E2(D=Mw^v@~HNAd)V}ScRqts zY>oB{ZkR29$C2&tzT2Dz;&EtQ-=8l7z1*P%V*5OR-#NGPc_;0L`wcg8D+6=9YcyC) zFsJiCUi0H-*O1>1V1trf+YeoPdu`&?1z~`~JE^tP{uIY%2>Wc$$79oNTUaOh;&)bx zfFvI_dmf&gI$-J|nA8iwoNn4}vw?xLiZ|O>W5Ggy+-@5xrLnNOq1w9(di!nB)*ktQ zhyic(TT8=g1;Aq#2E062=pu3eT1t%3Xn-LgH!H^20}NsNc8_fj!D}4qJhPvcV_$9$ zZGR7q(v*BakV1CGxgcnMqF;1a*n>l#jA5KG9B_r-t_e?oT1l;)W9WGj$nj~BPvcZw zD?{~v56!lPm2PK1eNeQ6PPWtTE9{5~KKNsijNOU$L;@mIhxA6M3wpe3Z#!7krbEQT zL1fVPm%iPe!GB^A`-#dPE!p(!{^$Gr`@+gY_ zEy<%ej{YF~N=wD~O<6=~T3=b29F-YM2&;XR!C)h4QAcs@mij>?d~VF@Tl;(iX_xPE zC{WfIyto7zNboeN&hQIVEf~qA${n=9#Ck0Z2JcT|ejO=VHK#j}jo**gqc>qq3kv#w z{i!|pUMvBpVXw*KlyjsIvbrv4JaAe|T~5sgh_1F`#eqY(B$XQn0CDY7Y+O>s ztjf-qL-}evmQ^uW(?1B@M~)E%aUjllw5?|nAF#PNx{@SJN;+!uc}s0)&3zZHbg21- zEXs`Oey^3SiLL&3qr<<*D-T&A+3KjtZ}0S6F;k(5#szrS&c60nN*z<)G-~93GHW3X zEj6Wh9KoY@(82Xy)pzGPuednLT$Y|0Z|Yv6{N$o^CXbIyaCTpZF1&kh#Q&Z9{H?eQ z!AECTc<{mX1C<~$=$u_FZN|wXpkbJ5#X0?Z@*iTrLuh(}K`e5Sdy#Q`330CH7R?v4 zxLS#8n(;mpGk&QuG69wdOtDCR^n<@`ci-p$jRc%V!Mu*WG+$eeQ`f6ELB;{B?KSd% zhCMys!y&w4qYEV02N2=kiYq8x{Q#WwqksQoA5qG`oI%@vLc{Ov-+#^G>{*M+?r+_y z_k3cGVXhC#9lTe>E7{kEJb??*a!cNfi<1lsPU3@_QC?;X{DBU765rFmt?G5T%Sx?IJEkz8p^K5bKZ$e|*AE;k~MlC+e( z2E%y`@0($>+qih5#(Od;f;@xoBye6ipMIeAoPE8DYU_Mp4qHz`?$yLKofXF25wg{g zvsH|>AlaT1Rxs;AXjsF zza*B)+i&^Rnk>)mybq6GD?!7%60PX zT~dpg%uRR!iDF_37G{**h@`C28>cKlL}3-`5cYLB=#B{oql8yV9JnJ995_kRFoiP> zqSTXArCQy8h1^J_5~Mnta9=nR!)Z)-fxsuk8{;f{*^MNIgm`peh>}1csV(q|0B-5) zj4)pR++8Zj0%s6^)P0eLs#KJAA7uQCfHB6igh}tk zj5nx!CTD~O^H(H!7w3$3$6q!t&2@5wma!TM!|9KbyVcW0v2_F$!#f=g-?-<7_tzR#}(&7_vVS zJ>~3$8x_?&J_nw=d#Rn>ax)21?&_=@lZ%8uF`iR81sxbg(=|Hd!q$Fn1(|y)ki3bs z`jV<pHJ$jK@S}#^zt)%s-dd{u&b4 zA^P^)aFGDlcPB`$rO4tOhcHI$LKE19a&7)ZiHd*eVlm#^--uzYRAs(>Kiz`K@#e{Y zZ98o6q&*C}e78&Wi?0&?ik1QA6Y%aGUEygKV z>jk|^qB>b{bKYdd=teR$I4ZV+k>D_k89K?3${`x&bYZv5Lz?BvjUu-Z4zteIFhv_? z8$?++JbUdc)Jx|t(79{(#wFmAt^IO;e4_{3Y_-4dB5DC&wrdWf6f?e(*ENTcF92E( ze7A53#O0gV8gAwZcshXeBloL>T9qml-@mux71>D!vx zeV9M@X(IQ%ipiH$vO1M84_d(RfsBT9VHYJE-8Laq3Vce8;%DlpbGK zX6gw<+tF83T=ws5o;6W}(8OndzSU0|HY&=UmE*h2P~<|Djbm*^*sXGRxxtTf68(=E zT0wELC;1qNyC3^m*zg33Tj7e6&0M4;ckMZw6?|i7@vLM5H+0~+GT|4O+}8qd=r8FT z-rejR&TpE2+qDlv`zzkcNGZQ5jUg|8b9frsEviwrqE%W57SON3{3h6cg49G+Ig;Ug zH4n5Hxtcd~hYs%vE?l8Abd=3j+2}ipCJy=#si`nO`?%;xFh*Z+k{jqEFN{lag?3*! zH)G~Cisw`s=g2H(UGJ8|Pi|i+_cvXz8G50JJIg!Ex(GfY1+9+rnf`eea31`AK_DCn zu)zcK?3LQRTu!@HB2l(~_jC=pWBz`!DfRJaF=nQBO>pxCsG~pJjvjStxsRC1xN}!r z=|ezkq4b^HEgOBMBou_|Qd^)Ffxt@1^2NJpHQn|*EA3=QOKiY!PKm6g% zJ~)pxOQKn34U5d9R6ZE_Ki8$zFMDSRcGuS0XddXNx7<|HO(~a>9;_OZ#jE{Ms-xxH z_%tb^Dlu0{o<(VY{1=ns)1-{7Y`sddGAgQk*{3yerc69%MP0bB_CVrcQ9thhg7)79 z-+j@iP8#0JnLPa@tD`y<*Vl7N`a7+~su=iICFeKp3p#7q&^*n8L{Oi@4?|8}Q5L!L zEXd=}{_53&NA)My2p$+x_XR~>h4<}IeLuN+Kw~~If!)gtG8Pk2Evo%Lp~o6XZA+y4`Ij`>!Pv(r zO|rU*9*Oc8xx7rOynyFJbUo(*a=BWY7g(gQ4lpT-3l=WVlFa^JEMzDDoyGY^&egpZ zeXj$B|KaAveHF2yXfcmUum9hRt8(MTEnK|>uuja zN7D^=(0`mp?rpxo%eLv^@BO&?<>sRo;{A=6BQqceSiMS$+Nss^+&Hnif21K-88vza za?k8;)$SKt=g8pfolCuKhJmC9B`-Gu|0LzOJ z`U>L%Za1@t*j zZPOw#2He5au#ueC_<$Vt@CTQICDtypv4Q6wR_iR!4FuhzM|`q^ z=5>#BkGxQ|CL zhfDML_SEA;K{khXWUbNem4&O&D1|)`KdkbK7((Q$DVc&KAHK?#4ouD=PevRXj(;v8 z;J}w{2n*RSP8{zv`p$PVobBJ$6(ugDPf(M|YF&V(NYiBE2MY6BoYx<5sk$Z+0tN$# zNe0HSAN!M-gD^y`ygL?Mr?qGeBZ!i+BhezNJ&&WRloySpqd>M|zfMq-RZ9 z{n$|_Jdw9e+JeU*8{(XISMdDy-Fx`^3I1PCoX}celYdJjQNKhpM%uzIt{3CZ)p5?v zc;Z%eiw7Ep4W(kpUZwa88GnrkIS~R;qf&Q1R)C69T&v-8kao`=J0&}qKa6LGO*8Po zS$YaLb(Rq>HkO&p9|I;=n_j47|*4nH**g4QpMUAXko!xZ6{=-@;NqM`I+$nOq<{9dTk@CA%A2EQ+1|CAU&Qs z1*@V~0J$ZAp`^~M8gUT{lP6Aq{M)WO0pdisDbzC%wVcyXWktbh`k;FrdoMGdz(vk5 z0NP9B0+Wqw64h6iGpGQg4XIPzmA%I!jG|bCFa_tjg9bgLn0^_US-6L#SsA$!Fn^T= zpg?p9R~_*rUQilWM>566o)mp??n&3L6n);Ehuv4;>R} z9u*tj=^SeI*!>Ws5bPXVJz6t`Lv-@Ab3mv$j4ygKE4=@`DRda8)Z&bsEYs!Vq1C&> z84V#uCvf_pfD?R927)9XxYV?#gO|WB@OnZ7XS+iFZ6_1HPIhN8Ts})tju+of@Yv%U zF^?i%m$CdO_J2{o2JxW#HnbOT7y5SI*SBAnU!Sp!-V8bT;M|jI zAw{h}08^HoDi)od6yJh?e=dj8*xkf(r#9Tky?Vond?C88Bo#rb9|CBPi%?2}EIpwv zAnLd}Cb9+jeKRo%H0sJc%sONYk&TXD@D+xnI8UxuGzo?NCTOoGJ~5%HQ=Em*zkg*r6FtSrFc<0~;Vj~lvsCE;Yo0~M z0AS6n09I?0QYSe;EGA9U_~z=<`|GQB@8JYAbsE3@i?pjejsO1c>PO#g;qRYUwT=II zeRa*fAAWlK={?J20UBEig1My5DqiCsE0e}Q{`98>M7$=+FitGAb_R+m0lffbU*vU7JmGndSfb zw3=cUr`c+D<;T~LQrMFF)3qdnzP*>C5zri_g2uF`y%b?=<*%~%7#=>8+HqtEK_eXm zlwsE1>(;*dwRq|-IF!rn%MRsoGk-h3fc_}cr>vyuF~`ff%Fr!vIZQ6=(^NrJGlisbzBKBD{xXI|(#pgi zRPT#X?~^zXw>O<#NlF*?l zr6~A0Y-I2wf2r8W2sk1X^mW(Lp_~H1?lyZ^x*MM<@+qLvTjb@JLj({@Nn*IXM%|v~ zU)Ov*X2yqcEA^FIf9wA0KY!xtwB_uuc6lu|%tF7>oAP{0xWPI}Yy5aTua$e? zrC;xY5=$6Q2*|*M&2BY)RGczQIS`OjXLA&z$dI2Aeyf`51uuIjYXGmml4)96N63ifx3^Z%rB2ZvbtR*t1q7J|N8#P(=W1gRmV+P=BxKNtD=aLq*xVA z7N>c-dVjt8G5TViM*Xe*?_~eRZoSEJ6J_@KFY8TS6w&+MKJUACP|W^a+dnSU9qjWh zryqQJ3k_}8ritz#^9VUx``>ZaNz2u);gwo$xmZg{KP72l--flRqX$ZQfX!06$ zf4!=zI4jE42Buu{bl$J)!VV;jcHJI2R;@iiGu@R<*x0|R8$ZGryAj$t+cNpJD_E%& z+>}LZGP}-M8aJt-KJ758ppn_KH*wOyMq%OCyb3#*2TW#jGr*$k-SjsmE20U?+~eQI zR1p?A@~$KqZ$lbw2h5h&2`ZPkgm<)mA<)*xbx~p46E{wMZHhAb2R3Z?$gadE${cR! zx8;FuaIKSYnXh@#UEEGH@{BOM3s?e>l`zNFy-wLp2?%W<(Y zEF@{umV4winU)6cmS<0d_#k9@M}#uTDD8?Y7h#0zWJ?)2O_nq{A>_<4w z2`8X7q8u9hzd&nm4ovKB#_+8IcHWIngENs8#`7d<3I zHEZ1%Y*$m-MS$_N>-m9S*eAO`b|J5@;qF?7sTTB#TYv#{yaNMvr}hRPaqp&2^a96f zcM6!7EB`x;%Z~Et7e&5G?eQ;v3WJ+uOq{2{hMG9zk!P`4+o>c;^lV)u(doK}<*`pU zU3)x@>$C>a#FXw=yO7&sKOCs>U^n5ZvMn3i8#nzGjn`BOF@0^ z3J)-|BV`L0f|cvH)ZACRQG8NmYltqm(=>W3Ty^YWCmnCuQT9(sk*ukIeQ58keNS<3 zH!>Hs!2z9pg5mVU#PT#_LYbcyn`NToK)*~HRDLUmmIiiBMTD;$sn7sx@DT+86s5$L zesi_GWjj2zaJ_{X2jf5rg-X7BaDeh?3cW>N>U$s;7EkJb9d2 z?xM#NTPn}8=*u2YIGmOP^{SCYRW5C6$73cEAbQ%QnVqv`muF&_C)mVV?!`9a%6edc!Eocg3h>t7vT!In>J;jC%o@XI19>bFX0 z?m*fk^|CNUaR#fL=h1s!P$+EILh#{{sobXG38Y8afPT8ry0B2lRMUc2m{v(WHOP%{ zRwHvuC_!B0oCUN5ln`C@N|x9+izag1ek{RnJJANR=nUO|LLy6vE@t|Ll_=`yTb}#= zyHC`ds(f5f8cg)cKE41Ira)k0WeuKI=%f4f5_p|KYC;Z`NKbg2lUgY>T=5snwC!Ilg+h{aJ?X^DN(cIM0@KcLdGG_7giK6>5{3(_tO8URC< zl+llc=WrQ+iboWRkcF6P?1irS-(;nd!mU&iFa)S2Iu_<#5wOSUpu$CE&M`kGEX+s_ z?tZ@xfE{9!@xy6tU@swM*E-NiZMOW}O{~ZMAET0QJ}1ignuX|_Zx@0555RJtNu3x+ zw}4=tF(hzmiG3__s6=F_BX1@i{)qr4p$pK7kaMDcJjaE@<^KSA0S&<`GB#3J50uig z8YP8)AGwAYEQDZk24pWSxv`#r=cpGMizFoKB9yx4O|C<&@0xP2RLSpvJ$Vg?`i0%F zOjeVTE!yuHa_u@?AjfXnka;iAjN_FBk;y?ukmTo3h0|mK?CTt@1!WePY~d1b1uFJ3 znf*k6)whlZ=U%j6#-KDPT)IVm)r}}ri-1Vs)velrv+rfuSidWb(+LU_PG6F3t m z?Q~X>o)dSna8@ycOf6o&!2JPEZVEhE18(nd6ql=dmSvy5j#AR929P^k} z*mrNf|Mmr8p=WQOzxwL$FW=GIn=e=^dG!+4$Ru%N0OVzHlXEN2zkSQ4H}v!R`*%No zWoi17-d^zUH|sP_qR(EvdHwq35yx)d<`#t-`v^!QZ{vv583n#De7ZWX3+}~UpZ8M&dr2h<|bpjxxg>+?Btg3PNLb1 zs3$IsNJQgvu3(}E5mCAE8*`7FihBO-9UG<~AGy}f8~qtx8Ir2p;&E8sicQv3f;y=w zv5sZU%uvyJ#(~6B8dk*Sk`9`G43sYWzQh_}XsJ4sxL4fk@^)PSyvl0n!2&2>cVc+y z{Q5$Mh;!7=@U&}t=uDJooXNU&U^s=vxPWyD*4gCgfTdApkP4smgCq*L$ z_ixsC0%7MY-m++rUyn3WLI7n+Etf^~+96HkFM!c@RN@&_#gsvrYLC6}p&83fO%+$Q zGyDq(*3cmE8oYe3{EO^6Ov}H$mcEm~+|Knuxgw|e*F zZ|q{!BBHRI-4;lTCeG@Az}{vD8k(E}3311#>B#WwFZS;#|H;9`&BnHETVK-2 zTyro7_m9|nt!F(#^`GRm)hGG;w~rS;oY%J%eolZB0~%xateZl|_e)X}7=Iw?rz)C8 ziblP!WKsMAtWf}OSWSU7qm(FGB=!%e0R7rEZZ5_ zZiJ$tBH0MmR68{6$EdLW;{5p5B0(wJ@ebw z3$Y$LjGy8YApZPE7lC=N6tJ3|twS}z0XSE;>NeLW>D@m-BwkQHeu|`IZqXYG(WE{F zZMrvCK?qL^!?{uY;*qxbv%Ia4MU&yJHpvTx?I_pna%_L39Dvq!3N7{i6T5q8%0wO$ zKN>B(kuJL&PXh2H?on zM;2V&a!~M(+A&ABXrSg1=>-v8*Uq|vH5$y%gSBRE6tJG@`bGs z2()IXoJct2EV1taAl-kc^&3a=^%p&X4sv z?&hTgb1%snYyy#};Mx0328I2ktkjeU0`WHaA{ZyV1ZL^anL5Q1QkZ@aLSUbkm4!DG zOw#_EgenHw8@n^Rcn-2K8-GJY8MEf^KCQD}W*8WjPWBBnZgYagWW3&U6E9Xx$xy_l6y~y?kH>cQ?{0 zoUS+0`0K<920__mAAq|+(6ZclbQ=anq;Wc!_^^LCEUig#qRlpef8%7AJZT&35N$1E z2j5%R$MvJk1gp*1$fRJW=V&%@itn~>DSrwy(kR#Kr>Bo7KoDm0t3;U1tH!@4%VxCm zZ_lOKby0z`9CT=tR<*cVyLA3C_R7$b-0Jn_y}FiDHY8|>70~fBR&k%s4s2l|J(ikW z(u>q7P=OHp^dx&2HWtU9Z9xGwrk$t7#~h7xijc(qL@u$*V<6}5wwj&B;gX`z0v*%- zY-E`Ih~T`TP`gPzqy&^fJ-RJbd9v6Bnux0)TL={%O#qH4!(TNjR3HSAcIjF>MKM`4 zM{AT?R{1_qA22#iX@Y(Z=R7}cIiL~H^OP!7nH!n?Pf!vT=>=}6=|ubEKavjrm^iRx zcUQ|b&HhoPNi05nP+REWkUHUW)MR+$#9K?f%-f&Lk_meCLJ<7{0d7OVx}{tZ!6rI! zSNI@D^_eMRGOCDWF`+HDe4gbj)7@;o;O4Q6%#xN6RDf=~`Rymg3heqIr(hNp0iM8q z4ACyvryT$Rjou|cC==Nd5Plv6P>);FJ_u7prLP) zD2;$;2dB28j4p+{pQE~*7261RWO(?rWDbe(mWROFad!Pm_G z{J^66{*H5NRY_*8du#mLS7!vRseleO?0v6Y2Am}{h_>@r-{*?MnRsDh+VLY?r_k7C zWTZEr%hyDyb@g)&KFBm(V~QpP2cX1zFg6NTc@Ur0qi-ynLz(quGY@k%m~lv1Fx#9? zE24iE4E**`WEaSBg8bogaI7+j8!SH%lEhz|83m%&q3TogVsdO0P!SN|z$IgNT#=%N z4(QtkVCfl&+ppq4O*Z`l*LhjRih^_y{qmIgjS7>FHJv|D>#RDzX=Vm{zyl8qs)9pC z?I4Jn(JalA!F+k06AY(OM(4XY&8=$Iij=~Wxbt`P5n(A=%OmI{`>J_*V1zQel=g?^5WHa%_?ovEa1mwyb`>ni44RvIM+j{85T`z32(0rL+oQtXb zp5Bo{bLbjTIabCrgsCq!IsHQ}1}LjDtYg7?Y}$zhE`$)e(iN`9q5D0mScxT8J8T$l zaR@>L%;km(GU9v-!h8Jt_$5XGHHmmg%RG@&`oDuuQ2X`shX&(>+ifi_Oaq2k!c)kCE6S z?=u)|uL68+H-t&ei1sdRm38v~z+n1vZfFq~gvaP&`s1Uc4<>9&=D!o-wzo}L3}PG0 zZlpITl+B<0)lXflG^qG+GXxxl_B5ii7v81q3M|N!y-j3X@!h-ncF|^2j1JDF?FMNh zrk9dAy!u)Ae{s*3;Uv7^P&R|??VmIR?x`aZNa`n!yJu(}#UwK<3?Uz|fq0rA1j%u~ z)(6eqksnwfU=yCw^ES=A=I-qIxv=nj^>AN)QoGQaM#U}cBaFoogUf=Y8P$_T3E3{e)@vJ2h}xnPgKWExLW z2rSes|LZ#EwTznEG%j@zlOMCsR&M0cW4T4FenRY{QE(KR(!^I2(K{J;q=pURCX_Ft zTZ13{Kp(8Z9<|kbONjDsD#{C;7YYN-CcFBE6{aJEd@?0ta1xYy5`ap;$&BdBZE1Sk zjP7)dWJ)*mIyp6FGIDuo>VKJl?AKgxqAH%nGL-!1a{{(kKP|JUEON3bT>A1jS=92A z$6I^M1`=zhFVCo=H?B02dKo3A^_f801M*Z_;S= zdYHXBMi1Fn>{%J14xoBZ=8b^lbj;jq%<-Rupys7R5_>hM{KlK5=QIkchkYC>xB>Q% zKBTHC7G%z{m40$au#tljfh*rQFAu6cFL#Bx#OBXp^=M{jD+?>!m2#b81+>yXZ!ZN+ z)$drTR`w&=Yl&7kVtC@`IV15uuqu2Glj_v++p_Pe|3+s0P6O;~NmQ!b&1XbM!mi$V zZu&NqDV(xT{7||7XejOQXthU91|u6z+Qm{my5eGIH^@YhO+V-&gpm@ZDI2h1vmzaa zU<**cbtRFW{7U(O)eJhU&w^)|$dL|<#L^Y_zV9L>VZhjUdzTz^Vm8s9+kKgp6IS4f$eURY~^7u5EK!D!G)|&A&^+JVGarQ4_URB#k z_>MWqt4KN1+g6c&m-^>io{A$qt;QK*yQj-oQFdn|4(aH}nSG32o|7BuaOT^cU6Ow> z#{Yzt@70)uAQat?c^qr=-2(n8~`$w3yp${}EqPOv`9-UgC9og7>q10bn2U z11e#zQ%K*CzoC3X{f71p{Ts$N%x_rVu)pDa!~KT$4gVX#H^Og3--y4Fd?Wov_Ko}- z#W%`tRNtt-(R`!*M)!^W8^broZ%nPOQ_Qb*5Zo-R?Ekm9B(UbZA)R>hi5cmoLpSgf zJj7>vv!!H%@Q1hE)urU)B)Ebo+p@qSh?U;wlbV-?IVhxV;|B?8XC2#3;J93wSGnzl^5n26l43xfL~|L z{qRF8vr#wj`Q~kvXnl8oe8XM4^Z9&ye8q_IiP@79wP9r6_PTWcp4-}MnErDK$E8`m zQ6p((@$w;++u?_H9huI5&tH6UST}y%UQY14LSKw>wdWdYDZJ5EG6#*lf4@r9Kb{xd zIxK5b+pmtce-K^mtbbiy?g;1ZJsJONd0bx~aX@YX#6H^UXQ*qO&fDJmwL(_-y#>l+ ziQ0lVwR*wZ{BAlgHdH>n>n&afo2{_o+VJGk^vq9^s_qj{9@CBcF&^H&-_J4cXf5OlX&XKGR*nD78 z#|N7M|Ge$2H&~|QbX#wz^))`UMZALK=7D46t%xv7+1|d;6-V&h+xibMr7K}`Jsr>2 zmjojy4V9FkmbF%(@6uLdl_ofi_($C0Y|E9cN4m_RO$yhl4l_(kadAHwE}sB0z9hM! zm~{v$D89I{_FB%5N1iUtWBtKEVL2Pa7j#A-;4@5#v6^vl6Sl!O)S59?R*4}?OOqU~ z?*eX_%q5!)7?!cFUelj% z6G21}MC}7Q{e6^3c0VXf;PUilHqL19)ziXrOKC;d%i`4kR{f2;*(8Xqos@#hB+ zI5FLAk12vvNR@mr?xeP~7@379Ff$VYjKe{F_%6o%EdZ~Suo39p4F>a3?pG>8UnMwO-)C(5L|7-nP$qvt8)!GL@11+0>O4ii0sk zDm=FI6Zri~WKe<_H{~ROQSHTs;D2h)+?ZW9_Ow&vlRE>>eo@; z_X@7aPv&u=G-FT~KkCj1pqPsp(ti5%6@o3;$a+vTlxS;aHU zOst<#$VtkLlIg z$zfoS3%iC7EU|+qU`NEYx^DkIU zAqe>tqggXG31c=A-`xNb^6K0|UMsz7t}v8fulix9|LTNbut{) zLCk3eGVQ1;wIWzrxJlUEY<$LfVPpRlT2ioAf4{z8Ij25S7W(VF^IC$;SO|NKdP0CG z`JkV1kqFrg*nx~pxgidQ$WnX}*&hPDtZ-)UCzv5F|7E(F_>SuObf6Io-vkg zK-jwRLbYex2^4`x(BVO)O;S>{2Qf01K%xPorERm%ve@Rxs|7t1hiWxO6GAxpu?deg@~JjSiH1YVjFBT zjktG|pJ&qik&}|Sg=mzF|NV(Tg`au#5A{!S3%+?N$Y`NZX;vLxeKUu3Wb}0R?)oiB zYKPu0XNv-8I`8&m!DvP5n9Jh)Z37=jaYRiKpBA;KYjeh97qAhX_umq4RrjKu&{Nrc zFq}39eimOQ?H^ay!uuANn^j^Lud_KNb`mnP5E(Rb!2h*1O1R$QW#r(DT`hjYaZjN# zKFh(9iP#E9{<$vRi{d}ezo!V;EMvy@!&p@5G@!_EndIr1{%YeR#M6u35hX&*{iLE^YfDv$3LkG7 zhxkC>pQm1|a`&hljqsu_PO8L`=jtNpZ%C>Qkqtss`@$2|? ztjrClzvivtIZaAW&@gVrI&1roGA(wVSxIr}oU`IxcYcl9_YV?fpY5hPLU(B{CJ8t~ zH&lUI^GJx_@DB>8{@*)&q%dTHFjnm|674KW?o4zO)(zSJQ z8#Q(229sBqRJG$KI5VeDNhIthuNHEGMzuO&d*NVEhnh!{E%AqwUA!Bh#*jKn==ef9 zhL$|&Ac_M={^)MZB4rz?n1Wro64BB05Oqb!2trGAkG=E>nDML-7vN9Yt(t04m-`EZ z^LRzdQ^-~Mc}n$xHsjM$R{f^5UKs;KJ3la%FgPC{nn|N(42d5RQ;Cwlo=1c2tFXgF z-+Em}ZTXYLxNc(7)pS(M%I9Y_mtm_f5+XHW$9q6puq^F&%1aI1)gRmMp`P~|1rH^V zQe+*YF4J1o&pccstJq#wiNv|}r?UYvj`%{y)yYay9T|*=@HDMuHTTlqC&g4>C^1^8 zc@{qg(e{>{rCvF%C6bn5ge8yZ2nclMu3WRJ)#@LF!_%-OEZIdp#7=ce%CAGvSt z#?NqEMq*3R8EUP%<4E9v3S9QxmHF=%{RN5>;c-;A5w1WXcn0dk?2^`L&GK~Nni}r) zm@fS+I_#+ElpXZT$)}SZ9*zbOGkN0S82MikPtA@*k@qdD1^Dj?B4dyX4in34elZM% z;g7w$7IY>ULjp8~OU)F`q%Q{*cJ~5Enk&oOo#%c6D-#O@Rp}IleA11fSVrMz@O4nW zYGkcqcJlQRKSCC6ACwR{UqT$y=S>u^%BWrA4Ltm!fZxW{O}Wi6uDljNgV!w$+ImR8 zSM#yGP@@dx?vGI#$V9DJaQY#;kKGs^`%#}7r$^2Z4Lz26pO@(dZ%rpsL1eQl0%`U} zur)GnHMI;eGL=n{@X34EP{J8m>{{(?yDlqqf-JSpt5)li zBp80l(3|MX5uGtga5Qjpam8u%=Dq{vM7Gkdo%8EwfexgUBl|cI06I*)jyy4Ru*-hh zq(fN}-`1E6K4UM`KR1Cd$B+6aPR|?iZ8P3%_3&wb*01>G+R^&Mj@xz=bCYm+6xB`! z$X+I6O?TGdMhs2;V_IXg0#pBn5hGTqsqkxnr%7vjtb|eG7O500W$4X&&usGPO2GDo zFY+RQx3PNY8kGuoC)5>PsJa42PQT^7ymFaUw_N!Sh7K^xYfKtfmOHn8zDz@tGGWaK zgV^7mpb*Y^#4R>Z6){}$Rju0)+KOF?*IDl-MC+RXA?fXXe=|3Tzg|K2SrRMNMe!8o zSI@Y#GCz@f|FO5Pud_uRS59L6mKg1p82$xWmZpDI^W#6@HWz~!*Yj7@PmT4I$%+&j zN!aZ=8Z;<}v_9642n+Z9n9s`3kA#0c6}^}@4z)0oaA$14KN5ftun$3)9q7={K708* z%bUxcoj1a9L$*IY(=*%%F98$7?qB54o>k)%*oP(!lEVDO#@BBfaj3m52JwqbFAe&b zRNcGdxlF_Wn%U#N5%UFy82f+p!-q-*)>$$Y^}_A9jeLv+3D_V5Ok$pk=#JA`!&D%@ zP=c516M8D|*jQCWfG3p**y;4;2G9@Oxt*0dTp`k(fNY&(O(oNnDWw1}4Wwyqe_Cwj zCMiSynEcxy@!@DOKD#xah&)B?nN@InhN%y3R61h;Q;xzP^g z6XmIpZV)A?asGvXkxsO;R0C|iFCJy}YsoN=CIGY;R_aDo{3ip)+?^FXzkI5ra;633 z>x%0n%DRFVQ`t$vu0J4h<1m22MW}^|^soK3WE$^|o54*_^P@6vB;fV_BBiGn`O%pH zO#U5+JDFxc63;CT{t`#&hZVCvjJKg`QOI?4A$2zQ9x&QrS?(fHLpu00KO|R1Eukkf zF~eO*?;;w$E|Ob^J+^6;);}~jC;lY^*RcWtr+4_$rC-VI5?XNbvzSaSBq49R9C}rp z4b)@{O=_f|P}8t0b=z@aK_RwlL5iseN;xzy{H^tKTq9p^kYZd!E@AEs)eEnwzYYd4 zI1W+aL*i5Je?1{6gLKs9xO>@G8Z_S2kz>>v5C)Lw{5|SvF}6uB5CUzf`Ul)B1vW<* zGJM?qez{dtF)>=PQoj=g@5P6-b8T6dBCaCJKjz16WLCsP`&@E?Ec$@a3GUIR>mMtNL`Ri4(EqK3@ z+iZ5D>c{ryID-XZYQMABS5_nCn{b@2Qgbbx59m(%MV?Kht%jume9WMi$i7x5av2yN z7?icti9btESeI%Q)F5OMLYjubt0{GZs^*l09mJX#mKL3yM=r9uIie83O+HZ7)EzZ8 zT_L=7PpoU@-yG_G!=O=CGhA_Y335r3J?!4!euhKI=3g>j05SX^_Zy+8(2%#=eBFWM2ZL=X6-kn$;HdeN1( zF48B(+R~a`j06t`VJ}K8tYTvyta;GQqRGB4HY6oMW>w5*4O!`HljgiHIEu%#eS)fX zLY9QZ)qkl-=c_e~x_Ay-1xGdxaPL7Y+Ih!w_7%$0h4WjOKe9zQTtqU`O2~I=lofSP90637@AdW4Cdw@Wl>>EZjIi8!9Lq3a(wX%tTvL5K>e;q|-4VO2Q4ZHx=4qlpt z<4~SOxL+I)U+~LACACR3}C(?(Jz5EY!-G2Ljzp*72ipQD%|d_}>_B3ZAum(8lC6c<`2B{I?-*kuD= zn9W5MRBUCP?ws!}-Yw1SJRsvJ=Fj>xbb6@i<3g+Y#-Kvpt_P=SeF^qLmU3+A1rG~u zJ47R&-7ZVg0-Zc$^vj8iYnsbu5h=T`Ej>-rU=iqr_>p{9Cu?aVyN5uSNw|g_z<-++ zk7T^l1*YBl`?e7&J6h|kLXlB~?Til7Aq8Oq{IiFlFQA%6RmV^00F|cI*i}bz>^~aK zc(C$y1A5g23V0APWrb?kHit1f)^NEXM=AWd)+)CG z`5YiLcEh+hMdW0$)oVO(01-nmXg%AkzDq~Om+|4H{ihD*$)sMO$Ml)DynTuW_P~&d zIK$yEmO4awxWM5+50CMr(T}8&>T|jFG0;3c>a%Ol+q(}tXAj4Zo}_VmkeU@-JD}s} zSG3R1{EQ7rZTu`?yGw3S$ut&34%Gv+W;tQ#Z^O{^CBkq&Iw-_Ia1Q?XRX_f>3>SBk(e#?igP{ zi_l3@M+3}wzK-9Tk7Q03oO*u*>}HE*oZrE?!Vulmh@4%VnF}IpIi4fEJgyVATmjsA z@HD7c#4RrM+!up;o~FvX0W`*|J5OFw>AlZRKATxDcK=-?Moy0noMxRE=jZaVTmRu~ za;h4sxHh7M^&_EK)U+wgHr)5{CV{zYUZ%nth#M;6GHD6W-_0+36lWpGZvDM!Y`ifY zq^|C{bE+J~2x$goA}0OXjZ}!>r~?>O7&jKej|eJHZ2vOPB$$?sYYH(Mir<<`@5ssp z+xw_#Cu7ditT-HCxv`RRRNF&xpl)Qi5~Px^CLar$O*f?U4bzIo55f6+OdFY~N7LqX z=-Ik>#wkaP-Z9zupJ+Hw^7*bQ*YnBYzTgGEH&OX+AvY`WC=u5)sqBx~>01E0?TSNC zc;YBgERMndib~4+5X*#10`1)dtzu!^*BP4^$y^W$+kJEL@VF|5fxnf&FlWf1aCfbm zhp%J*LdzYa6pfWff80WTA?mHAypsx?2;53#m8#P@UmxAcPr_-<8tMAyM*+*gs>kVf z70T}n$q&LHbI?vdb*3IfgA+3#OGh0zJ1$h!#^Kyl%|8rSv~LEsYJ;IwVzC$ zmGO;o6nvNmxH|mLOc9lbmCj9GDBa6bkl7frBJVz_lG8{dW9X-E+2PGc(e(MMHs=XcsQ47;?xw za)H7(?v2Twx5}ip5nZO_UMmST_8(&3tNI^ za)xmikWTfG&}T|-&-OcFD{6y#y51U3b|hqnyFgh$6%+7SFcUdiC)oV?UFnkFS(88Z zpLw+c7FD--6!w+A^pngG*8yV8vr2?Lyl)u@{p&1DyQLh2MOgP=(3rF z`OH}i|ZV?b8%He)Z-AW`h(Mf+)i3gH$>69wAZx6m+OO z;ljDk?-M{HhKpqG!k|uA(gWO-JNUk`lL$S7w|oDYs_fD>v(p*UsM@v{agn$*JR2gX#vsDLN~m z7S&I0y~?65lbn^>0J#?^XZBN0;&}Uxld%PbQ1HeVMNaSPlYk==SGeQF%yu(P-D&(P^~iZKu#_o#Aef%w*pB&e(Y2Y_PeB<&|TxL&$aT zz#-4w3KUN@9ta95ahHs}U!mr9?%npisR=0z-mz@`=l5c?)5AGjVc_9rAR_hD!H~IR zr_xa)cYe~Dsh6a+wxE!Z9oqfj0q!spZ{1k!Bh=9w^s(_A@07RC0!4Fs*x?%FZ#p}0 z`sF>o)_)PY0A;xwP~k_k`SF(UBd>4xQ{)Vk9~e9oEXvIfSs=07_0?=;!A%qGsj*-oB57MAKcz?7dtTn=(7`JRhS&AYqvub~t|9wsgZ^2_4a=n)i@$)5dWGW|J_H19NkSx&BNH?ah}s1S z%ku;@tW5U)>qSlEu|*R#{izvlb9e&sI?%JjobjCB-QpiIqMHXT?C2L;8R-KZ0B)NU zWubX&`x;aVrNux3Uy6Da==s0?D$p4-C9Be$INZr!BanGxdqMk^At!C6?5VtCNIB3m zRp{(@l_*YNW}UWZ_dF2i{;mbIRjjutf1@TZ4{ynf%-#3pH)W@iP=Sq3ownh7IRLhyb@%JtV<5**{Rfi=&N%wEmKEN z&?$ta&=}0sQ^rqnNuP1bdYldpg&zOX+flD=5lzvk&R{9q>BUg8B#v8_0L{6yYO{(@ z>bz<4T#AHf5$$vmNi4S+mf#`Z`mHrz);mw78C34gr;l{_2XP;!tTZ^b0q9NJ#Ga-< z8deDiuYY)9V+nf%p}f%i`u@X!U#L&g%E9p+53n^*p`MkrH$WoqawtOQ_TGb~VHN0; zo&Si1i7XD~s~~YZf!6{yTGRufl^ir_!;* zx0DEG0Ya7^NAi56CESKNE>EK}kfp6{QfFMEWy(n=bz71!mV6bwB2HHq6Of1a8*-V{GrH`0@=-)hLe6h!*! z;fa|IoXwh=0R=>@V?sL z;tJ0w6kn2}gzhpNEGR5+U1NX!8cEe`PS* zD;_${tzxBW_@B^g+hYC*Y8^EieO62eKoKpbb0|Jvnhi6{P!*P1XqB?|_n2oq%7luN zRcu`p0nuJ$n;Q@~wg}d3hz^sA-c^=`#e8w1j=2-RzF-|C$ccy~Zk~<~FH!iye<4h$ zVj4F$zQy=Qi^yvx@B}>=2fjP=O$||`(j){Zh?-eV zSQlibT>w4+qE{DPXHG%Bxeq(xHpB{LhCZJ67x3J^CBfi?IoVLEfcWl#cwRl_y*A4~ zQK*_1E?FPN9;VIcsw+j*3ySAM7$B&GMqxzuJHJaK(_W_PCQPF&WMysQG(Rso$k4P` zNUhuCG(Tf?Vy{9rH1WX?EX#E?ValLv1O3k?2x5ea^lzK&o!4R`bY5qkG^^K-p?O#L z3ZOpA7O}uuy{36y-AE&& zEl6)^P}R!%p;_-y!)o2lWJ;rW>{eTQP|9-kCe-?6g z9-M^AI+zd#DkO6UiJW5W5@N+7g3rp@BHvYn{XkJp7?Ei^jHGHr-7ZjeT9Rad0e;YX z{})ZpTX+VKZd^43&!p8{@Zk8gv*|O}4){@1E%IHD`N54Y!?oRfO4q`h9b*1uD4DCU zKdynnVa)*14s=TZVCah6k>`s=mgS`~(3C0Z@|y>=ILj=waZa(RUg&@Owb7a`H9wyp zBp+7KujynlnoAiAYOlw)=gXcd03BE|L&g8eV?$erp9e6mtVId)(@=b6Nyn(2`p z`c3Z=mdTp8L{2Q5DB?TkQbY)AM6 zXS>c)O;NFPuVcQq`HI29;yelrJ2wTE8nJnGhFeb5^b{w(Mx*{p^PL$(BD(=Up$kuIp`iR}ys@nF=VRAdI3+ zN3(zo7wO^EKZpcixCFaiDgB5Shg<64ZG>2}rMnkKTIu^|s85$fve3n+@@Wq>s zruX=4-ystioR2`R5W-<^oY!E7)g^ zOn?zrDB)rm`d)rNG|YLs?{d4;7(#-DUll*%0SD4+@nJWW75Iw=!!^qUL+G0&I*?%a zXQD$Rg-Zor>7fPcFWWZJ=DfI|O4JgXOMx)F^eO~Qa+>ZyMFNkOOeMk)mT*uriCYs1 zm~*AzPtkGYZN%n)KlP{`uDR$z7HSrx+NUk7{_?o2&JQQT$FJQc*@9MH_gQ*#=$$>* zs(5zK=oUsWn?T23?gDMiHU8Hx>}VmH7A!S#-?;KJE%}BZAKAt5ZS?M!DP3YVHI0U` z2&+xAB|$tvd9FxC*khBfn{^%{9IuDIP`^XEflrpv8OkD_+KB37QTHEU{pz%o5?*}r zFq3i_Mu9O35Z99X@)dzJeuW4Z*C30r^2YT$11P=!asVZxeuVb`#*rpR>K3A?qZiiX zq5lGA;^ovI*(sNMHROi6KfX|@kuIJ#SkHf`}Wz=Q#jCSo=vuaU6{V|trYP81MWjb>x zWpnJuK>}2JyoL=o!}`11XJOe*bJTn;wPDO@r40u)l98y5_>0|K`krv`#X3NeQsnTg(1bP@! zMh+DDniHM%b?`q47Mx9ov5wi~Sw+XEfS@t$(LOThO&RKqWm!Kw8RdM~%r*|x(RZon3P@R}z&~8u1-DPjEYzM$( zvN4A$Trs1spBQ1IXt)StINWC27Kl{{`IqB(9w7@2Z z?=w<#!MO;8b@bk{!g6ZV(QjsjqeY^b693L~ z>GXZ*1goP65j}d~#rb00ET56r4*?K%-6w_PuPSQEV}dE7)n@p5-Y7`-BEb!`&_@`l zbo_X9^IvCV2horH3FD0WG!VYADm(*;$sB|9W}#Z{c67In_Dp9PjtXQ*PNvgPtU6S1 ztpio*XQTL?_{hjKP+tu&5nwWPU~={402COe8IpELxyi9)@L?LYyz}pV2!A32 z%RM~(L*M8kd|dpjftau}U%>Px_Bzx@P_V%Ext_QQL2g~g(iY~OfUDN*(U5vT(l&{$ z64tB%v|7iVxu7vhR6OM8Sw%Fx9G{WGrHMY&R$n#24O36|uOvYt-y#=;eNqsw9tw5K zf)fkk!O{Aa26&D7UHL|wtGp#?VIpXO514to@q5nS!;|BRB_0&N5P?_A)P8V~NLo)2 zxa7y#e(>ahWi7Q8--!wlM24nx)$89vw%dfnvD7O|Uw=Y%%|&`Sr|1WDtQ!h4sp3uz zc$Q-udHbuXk*6mJ+$+|Xl1K*9E6pkbre-2QV8tU7|z20|?N-MRm z!H~yY`oNi+9^NUjG%(Z(%Tu(#4+-B9*NQ67DJw`UeST9!5{}$6wVseK&MVALD(XVy zBu*!Q6?qUDH4F@Q@z$izI;F4P^66s*``-9bn4gGx6yJSlIS>f@EEYW~*@mVfReZ{V z;aLyYgzizDi)Kn0PbF6PuMCWjt8A=|2&yW`FJpgD&j5>p+&#`*@V;8dtfO1ALA#>N za>Q0x8YDdUQekgA`?MktVr_x{5lRsBKOM9;{D3Vz=K)S5?+VMNaRlSTGv~1L)>sYD zc8GInm$dds3t!v;j9SN$uB8j%%+V;|N_nhXU9Y%MyEh-6gWV5oc=*qm;ayD#C(6WR ztTMiljz=!(IC1FAf5+lpWNcLippX@uV#Ry(iW9zBvcwf_=-({ik zY^$}#>VYA)-R+FGagoHDq^7{Z2**x0sd!L@<2h8AQ1pg?vZwdcDWU**5e3nG~=x%cCWzhf!MF z`D#@95@wlejpNWhSf~gwEF|=%F=QQe+<5t&5eq6`ZAzjINGFNQ==^{xVN8iug6@k{ zw|Xd3Wa|PY*NcJK8WKnb#gdZR4a$Ck#jk|jgM$ggdbm>o!?x2$?gw>486XS4uFZANxL(j-5 zy)Vpi5EE86<|G;(pAa#->jnSvJ$7&2^Icp_`hKid7qoUtI>Abu$Ex3TyvDCsj*bl8 zo!xov_7ysu-A%;s*N)uj98Er*)Y3NWc>r-0TptSxD$5jW$mb;oh3klm<6jW{LcU?L znSm8tYJZLU&Bp`gT*?(_C=p*CaUro)=ab?<&{uNoj~QxL)b&5X1xxJi9G^)K1CDAZ zc3DWzgB`j|A<*de?QUsR^X52{n2(9umt@i>h-he=q^aPTI3fQ2z5T%$7y|b<2L+f9 zt@)fk7L@Q%%B{6CQdx?yY1mEyow^R@*=Xo@@#U`v;WpL^t)Vc!TQ))Rr}9W&NTXD0Z@At^m*3St+HuJW8Un@0y`IlmwHL)tl$Y6-DtpyIa*T z@kMTfC@!kS-BDvmcx#(3#DoXo4Z5C+y#^m(V_2{7H@4gS`4px^icV%*nfHI|iZsSy zcy%o_%~XZqz#A>`{{T@yuD@gtj9!JY#6O{9^iSwO1Iq1vqG(G0D?n|quC@vidtW(! zQ_aLJ350D;`d7oYkw8S6+%3#?T_gb`i>!#^nuPN#Tw9sEIV>>|(>+@)v;f+9p}q9Q z-xTmUNC^0NEnY7QgR14HR`hwHV5(|yto2ROs%x_EwJPe3s;6MHw~HU&a!H@nPuS=p z{fS2;a?R2&URfKi3Ew8e!+*Sc^Z#N4T`d%cc(E{tc(F0Jc(F3t2m(1clW`Czf2~|u zbK6Lge)q3nOvJ`g5lt8<6s~#mwC&m6=xfL7*avq%P$WcQ%|i{4a?AVYxALmW0s+|% z8xC0j>dvan{PHNJi`^n!{PR`HzyERj>h=3PThvKYmUeM_zbNvgYU)K*)k$WW#qDx>t&E%84Ao3w98Fxg8{XtZGN|(j*GWy)4B>Tn;GX+) zC~x}%y}-moHA5e+&RwP3id^9+8Yl z5ZuT;iW)dPWkswcd%2^Y|NEvcmTkw!8d-2pB@^pj2SEj6jn=LDSj|a?P6#~^aX-4p z%>5Ym^9K|EjgZVu%su|K=52dqu?Yp9mSx0Vu4HI&aX_Yco1@(zY0w8sUF5`XW!J;W zIZrbZFN?P0LD7Acf7{H*{hj%x34ny zkS?-C(Iiz>F3QX%byY1khgW~TN=)65?j)V*eY0N=L-%`{nvJV;IwmmKhz{^n#Yg1*O>bP7Qa9!dPdKPyhRY7vg6$P*20NuXms?!uGCT4;cnc2&fbY0mELj zD8TX!F9&tjq@+S2$@*|}ttM8sG+yI5!Ok^%8WTS~fAaAZtk4ZYl`S>zxSNkG{7?W= zTXoW*#@=h2p31!dk#+91G~v!CR`n6a$Z;G}bPfRA2GQmYcO(2Tixq{ z_mB`!hewz+Fc#y+FjB>S0e=p? zLC>0na&1wexGpLst}~~vN^9K`aO(eZ)c}C}^bQ}~?0=h(@hfQ+ExT{;!SZ2P-jZgz z@98+9>eGVXy`o;|8EvEN!zRaT$Py&IC5hlrrg2iBf2JnCr$y?%{=@C+^1XY>c+z%J ze3nXX~!_+ce>a*nys02fBUH7k-rXD%i;mw$~lNYDCqv=#R%Q=wDA-T`hXOU*DZ~MG7oaf$eqESn6+lqe>8nb z%SBi3^69+az<0Bi ziIHfA=X>JAjaPx5&;@VlS9Z(EshVWdMs^&vdppqA(|s_GMl{sw2~7byuRxSzFs`)4 zG&V6RL^v}hQ|Y8wVHde6k~$ZYe<=?pQy!a4(kvKC^{k=9U9VK%6NZvAkg`Rd)=5#! z+euEDv61wXSdN{7qx)Ivwc__UQrFqQL3oHv!7q$`QW zYq{<@)E$psQFwU&+BLGV&LA(10Gk$$g9C1Oc&6-DW0S&o9Hcl*d^CmlXxy39b_23# zWU47dg4FYHXUe&eIXOWZe<&FGvS*@UaevDualmj^;@B)h&@39XKa-!%6{HC3lof~H ze&TO`iNm=2GT|$Nnuya{DWRzkRP^B{I3L7g>ORBjVKOi-i~~!Ds=o?08;efdmYv{- zPFzYr%)IhjV983%^T0oO=?}4YSx4R_*L>g!9z=hl5Y8*_*>{g;f8oHq3KnJ;vl(@U zJwJw;lZv}qN+U%xQ=+D788qUA|0CL7p ze*_aY8i!Ch?=3uvKH`%w1}%l-0>IFHEDRa`Zs%rM7Ou8yod6RpppeAf=2+XWK<_D$K7&`kG7XHsv`y(TY zKOoMLmZN+%tm0MSLDu5+sT&>x2?OU0;xIZ)3hf7PqLSzM+AvT$M&(!C1@5qQwfxM^ zm=#1NhseFM;`a9|EoEv>mpW6Ee`}3Pn^cz*uC6T=_&mAVJ7y*};v2*&g_Iaba~=s( zip=Jgf~xrWMQ8Frb$Su6r(VULLu-QMdc(VuLis$x>LpP*O2{L}rgbnpY0zPQj}7a@ zq%}1xlKjT?l%~sfPd8cWu*C(lb0}wLecaN+aJuXsIFYA2Ji3*juP9w&f1?(%e1@5P z9*{-G4R3s=5b}JHMJ@?=ZG$W=GVjKksOXbM&Q#Om!y+qDh`~p%L^FjNDkmqHvCj~+ zD>G#lwtYLquBj`5s$%q%)EWr~h$!2-20z<{8R0A^w&8SO?^K|w3@yMF0I!i?*Z!2P z^emYXc!Y$hp-85_Cg!P`e}0qx#VFW{4wxP9GfD;(e8jJETJgkhVpk1D&>*962w{iT z<1gFDXgk66Kz_^loD$SRr-V)5n(~K_t*Z=%#ybWi$7tsvsrqvf2vY;nwTCZZ8KAb zjRb@4m3=u8K;pNKKYEHUwfxAquuVU)oKJQzd9IIoZtRyEGtU@8|Tcj}Y6 zqn?$H#4ILew#_uke|*7U!Q8jcEo-~?QJ*^5gGX6|qV#t)iY+uL3V zc_G)>CCOEPM*;FPiripc1a>1d8@N7h`{#K8sA%OGZygw2oUMwPh2%TWByGrt9XSTC zYzlNvv|$I&>yS9RR|o9=)E~ae*r<`1zwdO+lOh%a6?@%hf3)wr&>lR=4$6$q$KFWM zAg8j(QqMWkA3K`GI|3t$sXxoe3i8|1vSmzavs=?Q(G!X{D6KLW&9)5YEj+8LEP-uV z_{lid|GPufoaNF+iLPxg2|-N}qb8ND_}#R+7^Js8>TYFCh-jE{LQ8ZycNDliG3=;ljwY!5H$M-jn~YTqbn250Zn z@+b~7I@Ip9_K9_m{^XhZN@cM#kRQW|buGF>NLJAx+tlNcWJ@l`r1fOnEdRy{R@w{P4@qU-1}>b2akf8j5k6QeJ4;L2i;PIQn}p!AmFbd|-N{ ze;_EmwD(jP=;JJ0m=ooEI002XgfyK%Yj2#kjP};r@sT1f=CjyQxQ9L~4{;_ZmD^(l zhRygX!lvXBJ5VhQGJw+toX&H{a3eySj+q#_Wg48L=jhfV;%P7r;Yx%^Y;D$lBsSd=7| z+S}Y&b)7BzIKJ(jD?hB)v{7Wje^rY4c4o;11_hNMY1bdLuBYT#^m>I{SSdd+ zS}fgxf7?F#>AJ~0^F<15ehw+V2hA9}i6QCrd-wSBzmTUVHe>|^w;>Imf3Ki8Nj)!qNQdzPE8heN1?f|{hXK~-YxGFHm$^~67KVN_bKt4P2Bg@gtUP~=smmfjO< z5dk$&DyR0`S8)QvM%37oToC~w5b-Z@%vM1{@kl|LY=e8;`FnmECfll>K-f@ULhtTT zA>)bVXd@BPp32=Lzw0)Hf6x{9RKA%PBj>W*Puxo|mh(Kmw~z;p-0!{KihbJJr3ejl zJl5wF<^Ioj;1H_+NjOA+MH~PxAv)z?^vNpW3Y5U4i@wGn;eD9$QDmXcOF8&}Z|w8H zCO$H$prH1;KnGf>Or*8ow`uLIjSahe;~W>n{SRxKn#dL}&P@5n-+Zt}7wR$EWPMN{}|z=qN!B;41H|E@rPI(mp2dfd`8 zUQs{B)y3G^OlHZw2)ljte?dHTT89SEGPefMGo4NYH8eGoaS$kf#avr+8##7pMHLsB?ooXlx2Q!adl8+QPtE3RaHkx+8kV59{e&{PLs(- zeQs{mpN0DL`p;H>eYzjhw3)QiSyoh&+vzN=CT&0Hi*nNHZre`%QH| z!*9@ewbH*z`9Kf-RefmSm45kJeXiAKMz3!B8-A^tP&>e_AJd|qEN0WcUA$H6nV^qNb3yTFw$^&ItF>_P^A``EP3<(zp;J>D&R^v38n z^jqv}US3SLEp6&j56*AG)k1?CdX9_k^yLb+DzL~ac!}DVJe{_6O$I+^2sz#6F$L8m6JAwEa3wC!2o2yT9p`GXYk(rTV_U?7kZgYhVQ%e8$_&RQQf9 zcz0IRO1zucb3;T~(e2;9+ClZK7J4$o)_R9$yd<&BUvQ{au~L?1lWwh>|KL@u45!di z(p)!z?~7@zBx9wY+_YPmKinT^cA8w8{=ddaE=kva5sT6=g7@@KP3acryyEp~wrOEv zTWo}^5nvj#164&^%B#^^RMjY98@SuSOB2mp&|Xt)Xs>p(<*G4r?=6!WZ_630epfB9 z=NHTS&E&gcx1!7~!`N4=`;%{|P(mwUMo|2eCxsyo?w7t}6qnxe)t$bXPp+)IWS{F*wP$v>s==44$VaSTmbB>)fEqWU4YD^AX%Q=y~vwa+fm}N;h zIYWIXH5A{T+57 zwWTlTEUsI+=-UuwB-!;wfEl^bFXPScg6c_sPdH87373&-kqeWcQ?jnaJa;?2dI5AUa`x;{}XS%4yqC!a1}e>i)j z^5j+OPq!tSYy^8{d1AFHtwJl-Wb$mU`g+^)^0hHN*%4or>EujULbk_c25hBUs?X7X zbM;3zdqbP4dOe421}W10c1nx>4v#95nT3%VB&F&z!><;sn_krF8SaSM3cDj|Qbc8v zsy&8@f8$q}>BADam9xl|uznH9n!mJU@^7TA7o>`M+TaNIoaECdQXUl*PsoTDZZ$k| zx0+2_2gaglCLdWr{(VPpYn8#yW0xm?RYUCBk^WK7Pp6jDOP+a$246oXD!mZ|y3hpv z^bwSSY7^^i*IyT!EYjlg_0S%ML;y8~PsSX{sC^ABT?<-IkdkE{#g!rCpJX7+UeN11 zB5)6zni${5Su(ZjiGZCT+}0spGWqWS0h6?h>Y{K+Xv7HjAE=o$anz)hqnzh|D>0!< zCQUwh&i18eGZdrLd}8fnd;*Y7CCW+#{u-y*G}jIRj`C68gkid2MhYBX1mpdzztYKE zTbr|UmR?QfQ)>JM8_w|SgB~g93l9sla_ZHGNWl9EtbW~3-8zXXOI=ARyHQ9Pw)j1L zd{OL(>cb;MF^RKCDG`T~0R=sOOX?y@lzw*fbi;c3?D*yJ+41}NF{oq{8hP)^qf!l& zMR`?V|6kOklD|1!4t|;CN^mbC+_TU0ePn!$eh|iqjY%c|at{*l$f^j3%AwUTbsr%<1o_ukTNmgG^=Wa>MKazzkB;510!GbXI?AA3D< zE6}qKY*JcstNVt?{J)SGZ zUcyEOmE98iix{Mp;|}e$7f+0Wk}x+XSrVm+zlZ@e3m+eA_pMZ}+*En@z|Np_2HG0+ zMDa4}U+V2F0$7iA3V4f$*z5dOkJV~Ug+I%$_8Gk4lG^D>U#|s!H-EH31&j`^HLIWX zbx7~zO%$g_Oo>sma%yaFVqNcNDe<1RdD?}KEvxcC&Jw>Z*zspq1>A4NAV&rSk5EvZCnma49ll`LiR~^q5FaRG6?*+2tet=IaIwUa=%qBi`ac=~21HqtSl0tCD470R-8$ zhHsW=)mvl?{iZv%*zHZw78g;~q@FGQsJZ6_vb+g_(VP=$=2qqRBxMp~e8HmMZH<{i zBaE+e)^MYLO{CeFrKSJB>zfSMuvU?Z&v?;@arFcG`f9klk?5XmgjJSsf@=2Z-6J_} zpC&Et)ueM)c^seL@Geq=lSN=|?V(*X>B#3+!Jbd0Ce^Uz1N!x&_L27LQ9UxcR~!2d z1Gs4&x)BizIR=3G4Ld_&4Yoj(_DlsDMaj-gsM#KWReQ^hpJp3msGEoI?7;+?h9%i^ z0q0u3YuR@cS#IftfH7;X1$BOd?JXyE0%ToTKdif!Al_gp18#W8)oH4zhz%|v6>G%d zB(`wY$9zk!`_~o~_?tC?{rY9NeH*@9SyQk91He$>_70z1@xpPaQ&$sVEU`Ab&M~$i zP=E}7XJBu(9A&pZ#7qp{c7;>cE!SvGLgVhN7D=$c-=B!ywQh?aYg8+0}PpDtp z(cU34Ut`v6A<`fHg5>xw4BQ$Gu}2Fy$Tui|DhoD?SmUa10i-gJH)-3}M20sDNrAC_ zV2`X7fcY3|i662PA*-F%QJxut_Eun6Jy3Z?PSq8Dn@kYIuk7#`*lS;?dyX;;T#piF zhXhyQh6Pph4nz!96QyZo`5tl0xQfc68HcKg4I$!OI8^#}L(eMVP&FhVvGviVg-f4* zVBxADSHB_-d*STyeX(CzMRiv5f-i(}d(Sbi+TK^BC#@^D{e=h!obA18!x2G+4hUFN8{E79k~bYVvVt7XYhS zEv(`Gh{^WBs9H~UFMT=mCUq^;Nvj+NRDmg~K$CMMMA;@iJU@=%&$ zK+VjJll4~PkFyiILy#enIfO&U?Skvqf>GAe(v0-`sdd86HiQ^)EZjzWp`0UshH^Fn zBi{;-4VAOr&Mz7&qCH(QF@b)apJNsU5}r@p?qO-A;K%&zx>S2|f&#*UmqA)}Y4S*& zDEJs?gZ8;z%GM`k685}MUx%D5MpQZ5uwK~p2mLHG0T5`hi=kiZx~q_UBge~CPvqBG z?@R`l8>`j+|6Hx4Dx)$lJYNfcGCu&hSg@pH)kr<0=CSP=on5#R{ej56joi>GA?M8i zp4e>JN)3C-$YpOm!5zr~q`sDhGdb`1SaEOeHDs@_iU~?R$bs$;EkGqEed)W4DOfO$ zpm|4j94L4SPe}M2Q8pJSKD<}R1!mUuPC8}fQosY(VDG350`lL7Mv#qv|FyE+Yye0+ zzA#lwLUn9M`I|hM{CU? zS{QkxdEj>)RlX{ib8k~r<*CIK;~f@%-$!_##&?Cd0$?0d zziEYk>DGA;z7rFW8poAhtuSk3ywGs4kQab>J~o!K5ro7y$i|)-eg%|#flT~0SO`!> zdFqD;RTib02h-%le}r+TG9J^PM8smus}fpB$lD&-E~d3-yI{bF;03R=L`P=}GBkSQ z^wU2GcI?SnGefX4r2JyYCT2c>9jLZuzEtF65G*VoK`H}*Xt$8Ud;t>H^r z)$o;lCo*E4EvoY(qJ71#59DYbRJ}*UYlGtzhTJ%2)pJu&%(OqsR`s}Aioo2VS#C# zh!i#fQThkw&1qt9+0zQtO|kwp$~U&T`F)9bb?XDZz+zG+H@{B-zp(juY3AXsSF)|= zd?WlouE+f6wc<^WeLhLZ#Hj33h_hOzPQnijt=om296xJ|6=>>_tFi6i)pg z=pYiMLJN9u)jFOUI!X{p z5Td;$Ys6a}M&sLGG(N5%ZZe)zH!@!x%FR2|uf6G*3{`Z>2Dc z4NS8YCEW{*1dxtd;#2yu><#aL;e=yt?_$`rSsAbp4tf0! zaaUoo3^H7^l@_oBuhJsG`9l83NQB5W4_KEeibsZ!msx3~p#CRieYvtp&tM#r*-Y>f zJ>9XCWSJ}yA(ksd%_yhj{Mu2@;{Hj0-%a9%Ywuwr2+fWj=O-3v6tW%VKGBiRTjw0w zkmz|hVJ6Srgfn4hF6#adMQ>D;y-$3ujH!_>)>d#im78_q@t&IKpZ!82QH|tTiv4Le zGw1zPMH

{FsXJ*!XOeCVmh<_ESh<{Mc6dKPm(Rl`{f>ooef+7ogJ17)x4zU(epL zkH^pECj1Vf%4rrQjW2OfLl63Q<<N!yS88?Twd}lL-(!v;So;2v9_2SwA+P zL7_kP87045q|KaN@sTzg$(%EP{Pk3@O;`>IuThEZp0oDfSh4k-<`02wX|1F*bW-3U zpe?|Oc|}!iJ}1`qa$<@*10Ll@lqE1T9-{ktN~f)6FNr}@Itpy6#UP5N7}T6TufPh^ zdrky~XOoZg5K_sIxJ2#)H~rZ@6!{E14P)JxqKaj6B~6G4GuDiUuWdGea}mOpvA3*m zm=cqVq=}47ocadWc?mE^-?DsSErTg!uqMu&^jl*t*Jda&b72Mh(rSUNH5T)WQ!Woy zo}6R8l<(Sa%s-zHn-iGnYj#1m8`g+RmIbOk4iLcEx(jn~a9ph;+&hy_fOj0)^HWHX zj?Y^Ji;6z-e^%(X0v20;h(lth{BH}yNtu*RNMf;^n~cf5!w0rl;gJz`^9XdC;jmz# zwo~tY&M+*b%33_TvX(xLaiv<&T6|_q5W>s1>(O-_@bdnoqB-q(OAMAY#{E&3XEsP6 zoLV1dpIQsFbIz#^*vmOw-w2P&*J5Al-VsjD#wyqqwO9r!C`$!uE?lL^N(HRlds; zD@WAy-9qHcIaBn1S%etkfQ9+jJFMh~C&lHW#^*EqzXPc6_K$qd<$qDy37E=3cEmx1 zr&^^uJopsg(OQLi_QZRLN0A??UHFKCIF_K#AG7ncLRvv(A zL#LIrPgrOE3Co1@?;cbBDb5N6%Aczc4;Fb;W@&&2lZSYJQ0uS$i{L|)LDUgxu$I3m_m)Z ztb77Bu_%%dM9PFVkjG#}z50V^Taya@#lYwoMw&iU-1CFJUvH4%jd!{+N9HWGJYzlg zj$s*Zau5%H?l^1`X6w5wlrYD##{?3UwBOm>!JK-1^G0}rK{%|6O#b4pd=z&(Ks3%J zq*r9w;cSqeg$g6mL|Gj*i6q0m|APu3juQ1i+~3&*&M~SYk2gYTl_HDkbd)Hql+sW0 zab1nJCeI3Dj*J+j5V)7eAWwDG((IqgAM$LSNuKypDTJjw_CHFV><{XlB5dJpJ6z-+ z-4Eibbt#kEuV5_u4_fwuqHm2Ixv$`)Nu#7m0t}H>p0UYX{ZWX_Ij{!fp`NEvjw@66 z`}pFg{{__{OO{R<_}y8`$3} z4V}zwZA1;7fZPBLpb0>K6le@!W(F{EadE+t0Yq%=+#Srnn>hg})s!`;XlQ8vW%+9h zz{vfd$fu{H`F9%t`R9iV(8|`%8ffG6N%22+Q~?43PG&%Wsks#pAflkCB_%Hbpp=kT z14saEfDWHUC^{QinHvLS&5eOJjzB7asjUOR>K_Y$v8|1X`Cp}fInsY}2{{4`0giS+ zWAjfppqnw!?k|WIU2K_l z<`Z=?G`DdCI04;%oc@Ys1O%9vJK9+px_`$0B(rlc|2qw5M{}F+|H^W}E z7bzQ4TL2T|KkO#XcK^g&fDV5j5anMzL-m=2p^2@Hl{>(H1ZWD&AaCpR84^JGKU10h ze^2E91&RMx1pZ%<_y14af7j^0T;l)V`~2_J;?7o9@`l!*4e*aS1Na;?hBkoDSp$#- z{55Q>3>^S}4IFdh|I5|T+T6TXXxZ$?gr3*VWekdWCDD?|M~oDrT1?%Vm8LM zCV%yjij$#@$!F{Rr{OQy*xAA1Q}Ms2!Dr?FT>pIufIv5(G3?5stuarKMOtRMQw?U$ z(732hUk?-bP_SJ|y=q~;UzKgQZ(7(kSGmD!bufLyB=>u(`@)Ic+nzM_{zwmo6-9Rn z@FEX?&zpp*H~;Gfc{>NqegK1h2}1g*MwI8}yzCZ)HanyG{?53PvVQR^6#is8iv#rK z1C<|%OTHf(>AfAIk8x^UEP4lW0|-h@!O1Ck;?cuB#lYzB&|vo;$lg6%n(SB@G9E)r zpX3vwBqw(RhueDNcgSZK4re%51W(lTH)j%me;eX7Taj9i*(YW{%J!0*z)4-Yu28WlwcnwL6ame)*@?Ia$9TFp6@+OV6T1Kafrr%xtateZ% zI0Ge1#+j7cwPCJ?p(d+YGmrQxrScknCY;K>u(MLrkuMHfLSELA)A&%{baymib)8Iq zZ+(%C7)AWiO9E8>=oW_Skz}!#+_35dG|N}3ATkt>_?+is;`=iPVdW|nQwIK%+O>Sf z;&O|XsB4ecIx`MQpGjr3F4R2KS#SLTsR~f9;Xe!y%}h;I`JzTQ-G?HZomfA!SJ|h7 zE7`#=60>=!1z(btZugHNG-9H!s)~1iV2~F8y78W3HPg4UW$FCIuk@HFA}#zz<@K1D z^;1iMv#KG`vSoNC)N-lVac1F42tEcQZE%vU^I*ht7N9tVBWRIDz8E^qSm>-6rd5RC6SZX!=eMdRXxz&w{}bP z;&^`Ue~yY!+~r&mxNyMcDPMn;DGW^VUrwPksw}^1X+|cH0n77y-RrU@CyU7BmIOE> zTcv#p|5zdBQbeER5z_2iS2omth;+LVHlpLmk;ZrD&vQZ3M?K`~b_8*8|75_2sU|T0 zV-5gr)g=xuwX2bucU8)uxmV&k?<-f!V$gAbd9S7!>l0oz^tb(36FTloI0hB97R-Zz z`UNjwlH+dH@G(220)(q$MpS#oPK=S|aJ~-7)kgk0b-Oowqtemlufh3$1?G5~X>L%h z@%h&JJR?V{cbUAcr>8|I;FyTwb)`JemwB_Cid5hEIH*IXH`TS^CidN|(vfvY5bgC4 z)kI*^ZxTdiiiiw@>KrCVLKJHM?XtWlpZ}MQh$Y{(y!TKghxdnLJ=z`!s2~_e1ah*I zC2KYMHz$TDj|osGBLUxk3|gTr2_suW59>-baRZBJaNYFtFc^Z2Cwb0LhCyBMuyx@V){e%prbTp#I>!hJ`RPr6G;A5_(>6)@4WQSj z_lcxiHU{^tC3+*z!{^TPPNZ=$HZ2Xt?^iY zPkEiB_?j4Ppj`p^M%IBUli>8dfEG}}Ide6ZUFbByxEy^3KwRUdbu%4x{T`)(H=gs+ zG0@A(t?vOf`c)-=d7UBQBzgaAKR0FrGhOQ0zvV~i2&OuPylLtKK69k=NG`mttcQbU zIHjBs@ACeZPnK5Ud%a3Q5Cc#(8**DBJ1)emr5+MSU@U+*;Lx?;G*kdY( z@As&jF*kGxqkRX@{`;lBUV)}8nNw%9WY)ADU$oo*=xR}gBN><5>^_A534bxN<#sw@ zqV8=@kxd_eIL5?=X2ku$lcrEtO4F^%HBR1$uUd`vyL+kDdxGBR5_Ym{-sB{6P?z34 zP02x0XoFKv!S4B$@btie2x6Ia9`$Ja3|}T+;A_eqzy}(Ge)+{pOFk1ffDaF}8UHN~ zKPwO{uBzyuWDCK6yZLwCAomFotL?V3`gsdDIlPfcFt;Yb`JTs{rChp{GMnA?x^ zOZF-S6i@8m1UwP?*%KTdyPw|mbrdjwRZ|vG0cV211BxnU>y$ESsw1Lchbx%1tdLC%fDednDmr z!Cu9Gu4_r*sf1V_j65!TH~~?Hkf^Y~wMY(6U3=c{D8KvlyYT&RPuM5#6#GduusD4y zLZs~YgvU2xPqiC{=&_o^7oMrSb2s>_{_9j{-?0D~?BnAE59W$IPfxR2fb2!!j%a*L z*hTIyyW321zJZe;V?Mst!6Ju^ZO!fn=pNsHm26K=)pjpfG95)fL|`A5Lf6`3_X^8! zEwtwsr+pg&6tmoie#i*MR`%k=mtf*C10y1V>K$iYlf*7RBJh=x9GV^e^!;)33|uN8 zB)LEbU;w`oq&8~_^e}PdW*A*v8O9vUD5C!|pGN!TR1{^SJ&^J^khC62SVD5YcfYP(Lv3P_dUEdG%Oia||BH^}n|G@J>1ZjONm#q~ybpDYk z^_0<9l18jhQpsRc1H>awKe}y_l50k?*5jqK1#@7_jOx00Ac9@YNMUa7;nEA0 z38ezb@S>-i8+C-A>SNp?ty|<;$t)xno|Rp`ig?OtP(wq%JD_^oy}zc63QSFZ-KPP_ zMI(-sQ`hm4Dy*Gvrx_x#onSO;s@J@_kk8Y50xLY$#X+Kzd6z-#j8`trn6q@NvR|Fo z(SJ(GhZ;h>E8sjX!`dtSbde)W$|bv08Dy#EAp*t13*)EPp^?56ynhOR{m=P!C>p4^_X7km%(0lE5}Q^e7X+VY@nT)Q&?9ER zNuA^iH|3b91;mX!LH&__-&ImM)I3Gw&Y=(ui#dxmG4vtAuIyQ&LVB;f(X(chwoFq6 zV(&_}LSW-DX(wB~52QBg+duO=2&Os}7$F?xSdwRhgMeqvx=@efC!rjFW7LFghEzBz z2`;#|eFiaom((gFuJX>?>iYxGqWf;paj|-n5Y`^O)5xC9PFO9ap|lL@+r_-N zefRz?_pKZiM%T!jDJLmf<2LyHkLqXH59Hq;T{-3(+jOyU0&)udJC{<)VaO|-uynVe z>GoXLSK{58DmkH}SJKUYtYD`!auR1EsX0&eD(2gTb||cqJ}KLfP?=a|emoD%kQJa6 z*o>1Bw?f2(-m&ks(7E6UcEGPTuS5(!J9`aHNxZV^A4RZVE8;Rnryhr?hw(>3rD3AvD*>%7K zSfvHIlxFGqQ88PiE)DV#TPO8dH#H5ntAeArqLVSRK|HM9iSWaBgs_}CPC)4Bufw~s zMoz1oH8mEnnksBnJaW04-Vk(dM?08Ni`uYf9i8unM*^C=-R}V3vNH&Vg(zo5{Ue3h zSWDi=UB5zgoS$odueZ68-1T{+?7F4yIYWulwcH2Px@$tuS;|v)<6S6pj;kkTCk|YHI8rBWh&*$4dA@FhRjMsK zP|zeoIJ^5{Z#71R7>=6=5&Be#x3y^2;XqZ&C!6%cdCG%4u8|dYDB~CTxn9aPEeGjw z`@JSi*Xj$t7Xi0(-#fk(W);4 z8S*3^HX}>zkDSoZKKi07?snz2-EIf1dgPEiXXi|q#bSvS3a4^4r@g63ZXJhs9(@NmG>);-BCLAcd ziYh?j`OZnBW(VzJ=yy1#GbvmjO%$>uLI1?jS9Ujd9#PeCD@kD(M`PIrs zfbSLK9~gmpkcpZHF%uNHB@B|#Nq+aqP~Y`;c&$Tq=F5B8A-C_{L^s)!ct1*A1e%1k zbmj(i4{G>P3mD7{t233otgd0etL#i$1N`p_w8fDugSFo~Zi?^cCA1M?wAVJ`G4dU) zRtHXI=NJ&2CIt42j`p5g3RR$WAyrh`IPe{RUdwj;IlLhZWQ0;zR1GUZVTA?E>jsOb zoNPIbb<}wO)VmPZJ_<1_1xH}?ws`J!zf}yXc}F?vdWiAWd<63K^|#ufxHKV$jhja3 z5io>i4mckP*)KNZ`R^cu$ERwK6}HHKLFX4;=pSRwY8>eGUDSKk_&UuhAud-{%U`-x z1|mi7F$zS!cUZ-4xW~&b{$5d6A*dDqj)thSp24V445va?in%F*X0f-eP?GTz&tC?s ze0ETgX(;g{)gJ-XPTj*jYd*!QOAKCJGkdaBapDQu!;q67^g#SuN}-qO@&Md_zNbk2 zQwg4@HN1MaxoJ%qWSL6n_;v=H$IEoY+VeCe1`eo^g5PRvZ}oQ+1Xl%--2h|Zk^1Jd zCp>v$EE{=-#9B<%k6M04C>hnB4I#3(i=ZRLi0lvx9lsvApRbhxOHpXhXY?9N)paq| z_u}=M6$$%+30Cd|(Q01SLy(StuAK1+*m8N%y!`1Ran?7KBO~LKAJE))4^YnT(Y~&WSlZEZoTq&CDh=!w)i?wnd9Pr*0tCa+Tuoq zol|foz_vz{iEZ1qCz#l_ZQIEo+qP}nwrv{|JGpc2sk*QCv8(obcklh}wR+14D1N&~ zIYiUKlJwI(rCf)FXU(cILE9Xg?F#27PmzsF7~c&Iyiq>&3kJY?O{g?@Ay0FVD2Z9N zzDj&Z0;8Gx3YUT!N*S%dPTvkvUH&fdhPf2C39cy7E**$t#0-7yFkgS2fj?l?m>-AX zTIS)?K0qE9XaTq{^aw{?PH|4+S0L-Z7N*}Je!dK5m*gu~=~Tq*iIq#op)g?(h2|2p z4ot7oUA^bCz2@FaG=!ynoomDg5$d4`-q_tQOM`9_R2puH!KRK{UXZ|dj0A?Ojb{q> z`8i?>+fxkUZCW=}4?$1qCO(CD-LNoMt+!eq&GMkAK>quXcd9eePpm(fhk!HZBQ(>g*d)mM&=EDFa zhtVvS0)Tg-Ptw4#o`ZlN?QccPell!M;lrzDX6BIJ)&jNlE_)>7$tbBm-1Csow++J` zLZS2X0#36*Ln!yG|MEFAPN<%61qo0=B_v)d(X^X_y#i9DD%y#>qUGGxWFgd`+`qxG z96LtlGsdA!L!8|vZ4}+sco3g7W*}aep|-I*oB*qcM>MS=y3&r>{t^PZqleeIULSFT zINVL0E;FCepw4eDVXd%w{LFO9HA-mJWAow8LVt;|OIv+mh9PH;6I1l_MsqI`c6W1p z{@Ld0gU!|Rr9x?J12*Nl7BXDo}ZLRa(bkkp~ z1_2cov4|59{7_E)@Sbmq;uU8#y~B$`VQ2Rrex-PqH@&>dE5&D6?{b5b%3ze270_=E zgGrQBkj)_dJIEq02YV?4ca$iJYNjCBpw2DVWX(`y0uh+)&CYkw5A@& zWQ%x<_^jvkAhGrF<{?ivtr_xn-z!h(aE5sOx12RCo5OfEkIOw){TStc8`6B4V1Pq6 zhik7v-WGo`mzi0Qi%oxT#>TShSh7>hfAVUXGySO63in^L%#|_WPIR#;rVT~wH9=*L zTO%XnTMskmeRY&U6ge zAP+Z_pzfNUQ!%4rbIHhwXO+eW;J<9W+LXK0u@Qlb+1+P#5#@jSW$0eRr`$5m{QWji z;Mhx>?u>pB9w(3Ol@=oN9xp{fXq8j2hvw3Qt!({HCt1nygAns?VsXq_1b6P3RKL6J zfb7MO6k?AvTqYg}jN1D*7Z6+)VqsIBhx$XQIDcCnh{RP-9*ijMu1^BiQ1@;~D4;qc zN$xp+`hFj$!EIwdBBbye=%~Iko~)l^XUlbWZAfh^w z-PzBZ5AJE07)cyY=>W!_+Tl}#W9ut)Z6wbT8=|h|T^ge1)J3VhMfkDf`{7BBo%1D) z=6pZh7Wqrg8E}+bxY01|;UX8XKVT>Oh^@2t10DHfnzVBT(RkY0Kq6K`Ts8URzqI_@ z9BtjfADRSStPLj8>%#pQGMf@V_iPNVq%BJ16Hh533)a^Xt^oTP@LY#vYM;FAP?D0# z8{XH*br@_ef8e~9$qrK`Xw%&;{(&d5NvRj$KudOIbJ437e4@#Ft_VYcDxac!U}lot zIF?4j`r;?i5CWe>R>X$mw$~k8pl2$N!qtC@(+nU!hCmV75VEJF6wNitOL6qxG}Xkx zX#`Msc`%iJhX97{4@*fZkdb2Em}o&LweGXup0rl6ssuBz96i6^NeUno(3FqsJaId$ zAl_SGQHdQG7DA8r6Z|0{&}(M0F(~-bE9y1$i~3XGUOZ4q+oo4_TKTQ(CnX$^5iXk? z!A4aKVTn>N=(i|)r&pLmq?QjzmFo=mK~6=Dg90_ky8%vF{KxY!Z-Y*OO&TB+YlL}; zGsvSCX^R;<9ZXy(VZ zz%JVvF0RgU@eS!cx)*zpNXb*Xl5L^laV zA<9KEwS>>ea65NZKpz&dQ^>*7{q8f7lZj?QTgjh7MR-1ZXw|^=^ae-wDrsnZHez1vb$o*N)aAq@lCE}_zc>SK1m8eKQ~=-v3(hU>}QM$T-Xq0OJyLAwcQBh zRXtgP1EWAW_+(Vzup30K+Tg&Db0hXeO-iRg-0`9`jbGRSoxQu`mwVJXw!=2&?NN#d zHlvo>vk6dyg=hWb;!nMmAlL7FeF!&_WPz59F9JQns)}Gt@Vu8HlrHRlC_@ZzJ_8`4 z@PH#fyQ1VjJQMT;wybnp^FdvbE}z9qW%5Kz{GIRXXI}w^Ao;;NZzEFDWc}kP*>Iiff-?Bl45z@mE|-G6AaPC8Ot(D(AxY%;l}RmKJ?9YQqbtMTP*XtV z{N3qp$mgUCX3xcPVXnw0t*ATp7Z}jZKITyiXPO~I%kdal%ebKc*^q2$NYFqAL;yBR zKu`8iXGrD#ExGmEM#u)`qF-`0Q`5Z-XlM8*r`bgVZ*dT1EaAV+> zT8S$#7Y2LKvU22?j!3&Oo-trPofUvizOGmD8x-VIendF0jPC%o6&;@`NIlC2F8-Zi z12j;pqo;~gEsSJe(*0)l#TtSta8h3e69Hr8u~>|>ecDT64V-i8H9wF6d;I}*Nv+`^ot2t( zftX!akf4PDvG5eQdP5wNcW{VO%1Xl+(tW@aEFXP{lZ|7RI@N{CKCl5X$?{*zj3e~1 z;6tLLK}4Qsi{w4T(gS1*9eUG{58~LuPQo08$C8wj^ATo)0;R2S<&GY0pe5AjiFKWGz zLXYq1-tVf&Q|Ui&GqZ-;*lv+lhRwe-&gqUrM?Wn^t~EB|&4Xz$EPYMFnz@!mk#GDG zrm1CbpfgCcI{WR*Zqic@6J3=l%S{vg(yyI6CTjEsDQh`fvtp07Ez<=pi8U9RV!@@& zTmrTulOg605&<8>o+|8>(N5}>e~DS^bDQ9m=xETXwd?oUo+7_U(w>Ic&6jkc*=~xu zgc4TzD37E(RZR5u@YI-tYuBX+qan!Nj4jGDZl*<}Tk1~jYUtYXICx!`-mGWUlp(Nf zJiEA6X9=RQwkQPzlBZ_{hvFR|bS@cC9XU+%k!p9d+X1ygJ=-3nTR2A&nl|5E{5-by zA@B&U_TiV+sKcjFZ9`t#i!-y#Wh8huP{lgytNEupCHXR{2fr^&+@a7%(DRLt(W%3o z$^CzQe5MeniZthmCk#mq}`aja)K9J+;t)`B2=TC z*an94+;vDey*|7{#QLk&Ub^s9b>*8umY_PFCIE*eo9&p5emY#-W8YNKVdhP%6IGR7 zAsLupn>oQ2#2$s}I=IjsxT8;ei8}O)%&-{+0oT9_$Eq1WSt|OWN9|Ci2Nq9~S{;fc zGr^Us$s!>WewII(IekfhGa(k+mUzD<`3 z!3RO@oK?DTxD7EbtP*0^2Shr4z8@%8rh2PuMwOi^ZYlzMxqx{L(q(?c+LAlcNPJlb} z^4Q{5S#zI4N_Kc)rr|zpr3^@gLKwf(Vgb|=E{xdul%4K6;t-lR#y}$T+J+Tp;B)K% z4FJ$SBXjtuPgPw-^WLlS_MASQEKJd8VY@=ty`a0qB4%&Tzub^K{(*?H*GdgC#EyCu zIPVy$lKv-$4a}fRWDf>@FAHs6EP!kWffm)6^uP;B(GMfQkYVfJ{fx>aV@I4TW=cGG zR9u)A+kmG|rGEjfcg3Oiv89pdxnCAR=L5eGcJ};_$eIB(ki{W!7012Rf6|Y;sc}~3NtoX3qB)~%7~_41b8{;g*t2% zX|wb+_aT#QD(HlCtKNbvl5gsM*XPx`Ys+fBap2#}u~`TDo;GJoYU_N@h{jh=_l#J*ukiJbNn<_n7E6wu zHpa7~jv7v;BbI8oti+O3%KGDH1AA)5%67AB?t*fT=JZu%^#|J6*Rj&UHa!1%dBa>_ zF!$_zf8<%Xxj*xp=W2p zNqR%ob)!r>j6D%MH7#KQ3wH0e=4|#iN1GWPa-b&?q+{d#mS$ffP)ry8_-UD;(hVl* zNjzuiG1h`h%HXc7zQH@q`D8*Nv=HUu1rgj+5R;FioR?gSVpFv1wQ*A1A{M@Ly_GAO z!{)~bF1@62S}=fa6d-??O}E+YckzB6md7?=E}CJKnLI2^2!$36tHoZn^UY}ckv&_l zYySFX&i6R?0OMR0Smfa<5AI1AtJHJ?54bwjcZd#p*g>3d8E0hd~ctDmgj?PTLhBFbR;@p zM2Np@8&@2G11hz!*VZy!PGWHOX9!tP29V~rPOKHcN5~<(T?rh`@Mk&CjsFyHKXC&0 zRAXe5_XD3REVV#D_cO+#C*+$m71K5P;eF5G=*uB6H{*EjfDumE(&|i#U52_`U8-m! z;YtmJz;|YU|C%?Kd8O^2IYS};`ve@U)GGBcmuZut2%uX>X4lS3*}ek$KAJ6(5ww+A zvV^kvD2=9uXqCbMuMxFUrRYdv@B9rVa4uL5Y4a4;0NTd!wvWP1(AQrH$xr6EJ<~@ zbO(&TBp}s1QEpyL`IrH(<~)oS4LL|6J`-NpweWZSF%$9S^lTJ(ee9Bp8CPSILGNsK z|HFRcu!PlDf@^jJ_kX7pSR86uQK#L~TuKeLxw&@7DL@f;0EGQ&UPn-9dDgQq8kb zJxFFUqHMF0sYPSJr3gVwL6$DZ&lG3%ueQAuT^K#gBM_J=*jqTO+>4mkLFgUa@8B~= z-)ODwa8wEg7T3=93t*tuXrD;uG5^5DS<@7r{PMLPi7LwpF2$nRhAqtf9Ui;bu=oYv zc>t^X@-f6XaB&{{9tiaj_=9$bIVFXc&Xn^=MITW7he9Vfv3h=cDTt1<_zGL_R9sqC z29jJSUBmQ!fdOtLax76`%llI?H8~eqlARF}@-3PxSrgccmc8JpOytl^Y7j5xr2}3J z7Y0R;mQ6=o1MorUb?wxu0ZGm?f@*d17@$yl)nKVRMKxp@QBbM5K=z8nE|?7>17s8A zu#T3;U=mihYg+^=244({&CTx6u2)(TDP=Hx4v$K|R7;iK>qGoPiw*I$F};UR2yx34 ziuCp*@x%1z@3Q&d&11|YVmKWlhzPSWOnxpbKTM!XJD-oMBve$l2OHE@@bZ0O8Gv;= z-I>QKskbcivI1lgC_G$8RBgx!qWB}Rqx^4$CBdHjo z3irx-XS?U#o)0MW7kg`r^sE5xOF)C*w8T5Ao_uCU0e%)_waDU)bT1+aB=cWYgi`4w z-eWU>Qgxk%;@2FeeW%Ylu$xMu|66r8Cn?4WD=glTOCu5x`YF#;w>8 zjE;C}B(sgLt^}5SoJRt)l%Y9}Xw5-Py8f}@lvFD?qoxu2iEb5~Anfk?aB{k|p zvFW2+QE6M{t=D^033!g(fX72KRPfM^7cY!WV0_2lNZ#L=2;PQ83D#43IoY`6K9=tP zpy@2#S`@)n1Q0T?)Y{KtmYMMQVpwHemVfBe9aOY=5ng0h5xW{sVHw|XH2;C(E+q~E zP_p7~@RParrk&CHSe8POkW+{S+_OY1Y_S^h_K?43raE^HKe=n6r?!BXar#zr%2 zWAP5VC7diGx2UD&0+WCkWVQg0eju;aeYCL;{0hVQr0{h>=kblnJ7l&`FUO8N+mv?2+u42M>-8eg%SO zX{?xfK3D_5&tGx-PBC6}ppqUp)~yJL=HwbPt6yWiEuLrd-`sXJBYU=ot_m0Wo1)ic zfGPfRf)yjx0#vV7=qyRY_+L6~gLc)~T8YJvhrc`)A4o#4bw(>`oaYe`Cy;$~Qnd;& zZ$!4V4QqvCS(L<0r2Z^4`!ZS^kWPZN+;}NojTC- z!>bSc#7_Ul`z?Kge!!cqRypTJywfU|Y^H1NjTFN{eY^P*_f;a>Wz~T$f30H&>D(Hh ztW2l7n{ng?_G5(#%+7-pRCKyRE+1613RLNu4F5W+Z6byu$L5K-{Qx~kh?sxGbwhFx zD+qV<0DKjVG}Jdg$`NY+JlKhEfZ09inU!i+mgAz2$-R(zqd0;TLRrkf$l@vOZ5yQ9 zUY(~o_kXkZ%EeGC32OOfg+B#kp=1L3%{yN9zE_aNj>y%d(?aE{GFP^>F);!*%_SlJ ztn0l9Tzy~BOMRtDSz?>Vt*dXh2WvE&A*iCP1H}L7Q34llK8P!Qz#P)c0uuZS;4iLA zj6?65r3pxUjtMhtB+3O!jFL3cfG8XYz*^x^MYWPGb%dm|r<+*ZG^{gNHdm3UNUF@K zjr!a6n`JY`Hx}!A-898@v z0&3r(-u}I*YOAWxEj%3^-1{Z}EKM~o-e z-Aa#bL|1Tt-aIOLgOlkv98?X0u+|}(FnxJi?I zcz-+Ay^ejsxX{_I^k75qV>mbdDn<2N0yqmH;fLcM0PRR(q~H;yUGFey{v%@sFDG$2 z^TmX*WUpjrw}J);e##Tvd6yRYhNXIvF+9BDkkH6_Td`A7*)(~{`l7vkC_GnRa?Xz8 zo~e=w$1D@~t$u;+f=%1taI@(Bqi8)o82}0j#=<_v379?ObWpL3VrSe+K2l1=0W!GC z*HfhP07em7%L6>a3u1+U3s0C?yzW*!K_T1Pjn{RD6wg!cX_=}J&)HWhXYTOmhY_ZA z0y?WKEznNMNc)>}7RHgh38gH%DA)k5jBh)P$`JFsHQ72 ztv@LWg3EK;y{uR+ZIDK$_t4+Yttl7Ub7)q+s>IYSQM)5M67 z8_`sAeo?e2syhI^w(UwurfnqM5M4Ky3ak2%-wf76IC(N8)}UL z&NvASTju4!xzOV#d52-R05)LmB_-QNOi^c>&hj?+jvyR@yQuLVoGudu=g8;0{p?XAKa400@*KF-vrXq; z=xD`Zm4MjAhqw_FfJ?e|o*_9X1U2huBXe-iF@d}xHBB@1Ep!s(3H>+;$C$m*)fEJ3 z$V4(=SkLAeic`l6ZHMyAxb@WM5Gy6!XO0O;d9CIy%86pV3UmANV9Bx&-2BeCQFX)QY(IQ7 zd7zrQg{w*#IYsA_UMMs|_2$9>2`ULDt!aib?c^a>?Xh0HaHC}~=3KqI)9i@m5v`EPbtJKuYGltoA@ zHcz3)LrrIT{Pz!=J*w;6fEH@f!;2hO@#igi#K*HH$wLPAs59N(UluyvW@y@W6#24n zhcFg9O*4r-oR@CmCStAZM;fl-4ZDO9AyXolUZ+QR=sER*?k2F67q;7W~23qs&+XOwv}oU^FYoEXrUel+r{_Q@Hh*zZ(eFMX+Cu55J{ z28wX7ic&0rN+1p(v{adHenoSzq(I0?1Md1F@+nw^boq_*eF^^GTy=TU;7a0f(R~F= zs%Gf#r-Y5k=?S#A>iXjV!%pxNc`|VI1^ebcWJr)IZ1^WGX^VW6vLR}D^fKa8MvoZJ zLfB+TuznASW;wJolG%KH&>dz@M`f(wn>;;h|8H3JVZJiQNEds*6vrzNGLyPiej&d! z*rq|yv${ddC`4=-;0*nQf_ToTxL>5LjQrU0L^IW84TOZKnCAQde}&{c`)5_KO#SFz zh5BG;<$-DSeRelfaBlK*s8tYCR;GQ7_Go^jLusINPCy|mzoucq<`hD@E^&eKkZ_cb z@IZlBWQ~W?ffKUze9z+^Zd*k}Wj$9#La(Gp9t?SMO@1+YZxEFZTRU=GW3T`d>H0|2~z|+pgxNTKeTcAf_@H=Qa zo2HoiwS3o-73JpjK>T6xlACql2~=lg0``&K0FZx2ICJB{;#oTO-stS61J8@vu?KI+ z&bsZsz1f8eP3^wJ^x*2HGQ;|O;Q3N>RggC0|Ba$U?$CTjx04*dd?k0p$W461tS)GKT4>p4VG@vZe^T+&4iqbn}fYe@f;KlTP zg{h0awBeu6LG^V%038e?9}NFcA6U@CrdlEGUfl!G?%qJc{xF;hAzJIkL}+kP^3A?; z441vU=~k)5o;0X1^w=1HLHHl;xUB#D+RCU|H$*K))?(caczuOuuBub+@%`GMtN$2t z867?V2usb)&aYDL#4lkl-%Kb0y0AZzGj(Hq!bU#Dy4ss$tfjta72JK|Y$EmLk_73h z<)4z!E_m@w>yo8objv8r*dWsU54F_!U~cT!zlsew1^fm`ZJ^-lh?R>uqoH(^bS5$7 zuu^n@EEa@MzAC7aL!I zJv+oR_tEvZ!-(OftDP!mSKQ*^VPjzHpWMRE&EMHko<^~x`n%;YV^+7!gGp;b0RxYWj;!E7}Bik<7qg03E@R&nS}d)lFF4?Fc|y_i~HGSt^Bj6xXs zWN`cBMy$U#tRaR2zhM^pzR^so^}5sn}6k zR-97UYImJ%?HF6t((rE=B5!r=X&;M-#`a;D{%*ZJQOKg=;aPnfVE9RU>PFhRTOBMeP4*vjsDbIE zZft@s4lf9uv9zVoR8D^ZK<)ksh7%HWSnnzT)qwmZZ4kPg&U#6~ZibH)0yT-3gS1$I#U!U`t2f)h8c8>D#dR z*7F(AO%%V9`B$^CrZuVsxy_3v)ah51kI7TOrxQ}ZFjtn|K}|jYo_)p97xk18w~5a$ zKBS3|>M5O#Wo56b3~@Ap>4i#Wd$L&{p>h7UK1fA>tFr$LuseUKqks%m{8^nTg7HxY zpHt}vi8}lxne`_Gw+Qj$FqtD@rh{ZA&}j(d(&+}Ikei~4X!$kC93_(Vbv|Q?r2K=ll=E347f^i`9L^>O7Pa!9QfrW#_IvEi}-x3g+ zMqH>mP3ae`0ZI(rG+^}(2Q?v`C8BQ=S4xe1cvo?q;Ny*C57ed5kWJAshneR3)x$%S z;wq2xRXVMuv61eH&^+I_*n8t3-|TH5Jw=40(e zdBVQR?3nBql&oZpB++XCu`dy^VWz}|Ha7Z&Ie@dE&e0!Mk3bQ&nN!&?>O?_<`bY!G z+uWUo=1VPIcJ56`Y?+`pDjdSJ0=0KSLa~EvsozJ~EhA5@Y|tTJkBvU*A&zwPeJn_- zg{fL-X(bz7|iVU8M8QdMlrzxsDZ zUS-UZp@S7F-7ZfYQ{uD4JSyTGxE#)>W*fv@sRJwA`sDFw)}T?Qrtjag{@9G)+8bJm z8&4@4bDr4<+238Jh#aqvBr(9*clI;sUVw#znHi~&po!L%g#uD8Xrx_*;Zn=FUC0+$ zhzG1hBX*2|Qu}d+!Km4TtvIIlrj0POav_1?>odBHZEu3zTJDZ#?b)f2!*aZUQ^QW0jy;yjGqC=r> zMbf!tbaLD%sgU(yLBj0G=hXsO9HsP@v?xW5Woi|Po+ab?%rVD?kIngW|HR^H9uLG2 z;;LMcG&XWiu!F6Yq590Q2V1+J;~TIrrwXm2d;#o~NP_s10327V_+gcV1Hf(PPMmh~ z3t0b4kFjK29(Gfi*&+jJA<*ym-?XJ?wM;yG3^PWaF0Fu|ssf_JB;cRr2L zf*BIQEIq!6j(@xl%_;LjD2I9POxjPLrw{_gK z`lC3Nog=tzRU!ADxNi(=W^X3j4Qx>;VCO0nQQzyn6;Jc2hWvuS#QDhBMZUVY!|TiC z@bL8v?x&v6TOmiuBZ?=fRpp{3t-$0?M*(ur@%CZM%mAJ$w92&e3h23^_h{CNl4R&J z1e*#OyXVn>f()uX!g-VZ?lx4hKB}Gl)8+f{4a?~8Y!DsmgcK(6;^+tr%|Ev2on`$3 z(~VVO^bBS=57*uWk~}@0=jNj$Jzj>pI~|}HFUmv*?lR7Ce>1vyyVLUyl#tu%;g(c*= z+ugM$WRufsj?!IeKU>4^!#+=D#HMp_eNYKA0~l)Vf+o;ha&ZTq0o5Mod5tf)Cm!}B za761KB>5DSnY2p1-*%UmHv1ZHvNtWm@wY)e(8cpz`>6YR0AxBkdvI`HM#r|aPz?54 zIx&(o&Awn){ocIc$o=g@;9k_rta0w1;d88PZLc_=u?uA@_K`Q$~jplqvPK6- zVE8Aau~;*lrWVk$$Fb=*F@!O)YLRCXiy^4xYm+Q008Gcd{`RKAP!pnil?fT#VGZg* zo4511ihQj>ee(gU;eZL~&H3qZsuxV9bkF2tI%Qw_+G)<0(^;k1A(sd-C&#LQj%J@P z(~dXBRrbFFmG_jyVY7tG1;XEM=lb(NG6fBkowC!13=H0!TxqKgcIwAJ@i2Cp!B1k$pC7W zxmfcjvOIBwz?{QeDsyhc9!(;XBCduP!Xg1(sz}Lca6o(RHUCZ>WEOyi zdRU%iQ@^?9cVK_C%!lO;7W4aNP|5TFWY;E`o;8-g&Gc-x4K1J=@ugMs!X}Sc>RPE( z>>65vtaeN#Dzkm-LLO>?+%qdqsco)rkT-uzVOTz|2k^*kB6qFs{~8Ug9s@m~z7+n- z%{&D72Vm?oEd0O=w0Fu`k)PTmTKy1tIU3<-^p7C$i^4la{#t>-E2k`v> zuM4|L{I{7!zs#G*q~x z7%Fx!=sco=Bq}ypG@`UfalE;TB+=-@DYb7DVQ)}0#TI%vmp5xDHeI5 z{q6HpyhYdpnYd02r1VF|H@cJ|VUA07pmCIRJml134Ss?I?hpV_;uJPC=&XGv?RUPl zBuOeknpo7QoqI-vcJd}b19x&PPJ>78S!*(P|6QCcz(bm8OjOrQ38}4f$%fHC7D9;1 zJC`ozh{*YThD+#EM9_bSCh&TypdYPP%a;&4H=QtprN|F0fpj-8BJ!juEh$b#6*!GQ zCj>?n`c!;gMXF+5l(7`2t0;95mb;V)z960<&SF6Xtj?jkixHIl@gooE>Y_yh+o@Jz zV!*K3guBMkn*uZHYu-GNUb|u$==55?41fBhT?sPt^7E62QuCrw33DT;UWs$FQ?81T lPUd89=LULnneqvY)+{KKYP-r#0K>|}!~{)7CL$*a{XfA|J0bu8 diff --git a/Conferences/2014_DLFM/dlfm2014_Telemeta.tex b/Conferences/2014_DLFM/dlfm2014_Telemeta.tex index 49a8668..bd1da87 100644 --- a/Conferences/2014_DLFM/dlfm2014_Telemeta.tex +++ b/Conferences/2014_DLFM/dlfm2014_Telemeta.tex @@ -144,27 +144,28 @@ Claude Barras (Université Paris-Sud / CNRS-LIMSI - Orsay, France - email: {\tex \maketitle % \begin{abstract} -The audio archives of the CNRS-Musée de l’Homme are among the most important collections of ethnomusicological recordings in Europe. Yet, as the number of collections increase and as new audio technologies arise, questions linked to the preservation, the archiving and the availability of these audio materials have arisen. With this in mind, since 2007, ethnomusicologists and engineers have joined their efforts to develop a scalable and collaborative web platform for managing and increasing access to digitized sound archives. This web platform is based on Telemeta, an open-source web audio framework dedicated to digital sound archives. Since 2011, the Telemeta framework has been deployed to hold the platform of the CNRS-Musée de l’Homme’s audio archives, which are managed by the \CREM. This framework focuses on the enhanced and collaborative user experience in accessing audio items and their associated metadata. The architecture of Telemeta relies on TimeSide, an open-source audio processing framework written in Python and JavaScript languages, which provides decoding, encoding and streaming capabilities together with a smart embeddable HTML audio player. TimeSide can also produce various automatic annotation, segmentation and musicological analysis that have been developed in the interdisciplinary research project called DIADEMS. Furthermore it includes a set of audio analysis plug-ins and wraps several audio features extraction libraries. This paper introduces the Telemeta framework and discuss how, experimenting with this advanced database for ethnomusicology through the DIADEMS project, cutting-edge tools are being implemented to fit and encourage new ways to relate to sound libraries. +The audio archives of the CNRS-Musée de l’Homme are among the most important collections of ethnomusicological recordings in Europe. As the number of collections increase and as new audio technologies arise, questions about the best approaches for preserving, archiving and providing public access to these materials have arisen. With this in mind, ethnomusicologists and engineers have joined efforts since 2007 to develop a scalable and collaborative web platform for managing and increasing access to digitized sound archives. This web platform is based on Telemeta, an open-source web audio framework dedicated to digital sound archives. Since 2011, the Telemeta framework has been deployed to hold the platform of the CNRS-Musée de l’Homme’s audio archives, which are managed by the \CREM. This framework focuses on facilitating collaboration and user access to audio items and their associated metadata. The architecture of Telemeta relies on TimeSide, an open-source audio processing framework written in Python and JavaScript languages, which provides decoding, encoding and streaming capabilities together with a smart embeddable HTML audio player. TimeSide can also produce various automatic annotation, segmentation and musicological analysis that have been developed in the interdisciplinary research project called DIADEMS. Furthermore it includes a set of audio analysis plug-ins and wraps several audio feature extraction libraries. +This paper introduces the Telemeta framework and discusses how cutting-edge tools are being implemented that provide new ways to archive and analyze sound libraries. \end{abstract} \section{Introduction}\label{sec:intro} -In social sciences, as very large scientific databases become available and rapidly increase by both number and volume, their management thus rises new fundamental questions as well as new research challenges. -In anthropology, ethnomusicology and linguistics, researchers work on multiple kinds of multimedia documents such as photos, videos and sound recordings. The need to preserve and to easily access, visualize and annotate such materials is problematic given their diverse formats, sources and the increasing quantity of data. +As very large social science databases become available and rapidly increase in size, their management rises new fundamental questions and research challenges. +In anthropology, ethnomusicology and linguistics, researchers work on multiple kinds of multimedia documents such as photos, videos and sound recordings. The need to preserve and easily access, visualize and annotate such materials is problematic given their diverse formats, sources and the increasing quantity of data. %With this in mind, several laboratories\footnote{The Research Center on Ethnomusicology (CREM), the Musical Acoustics Laboratory (LAM, UMR 7190) and the sound archives of the Mediterranean House of Human Sciences (MMHS)} involved in ethnomusicological research have been working together on that issue. - In the context of ethnomusicological research, the \CREM (CREM) and Parisson, a company specialized in big music data projects, have been developing an innovative, collaborative and interdisciplinary open-source web-based multimedia platform since 2007. - This platform, Telemeta is designed to fit the professional requirements from both sound archivists, researchers and musicians to work together on huge amounts of music data. The first prototype of this platform has been online since 2010 and is now fully operational and is used on a daily basis for ethnomusicological studies since 2011. + Since 2007, the \CREM (CREM) and Parisson, a company specialized in big music data projects, have been developing an innovative, collaborative, interdisciplinary open-source web-based multimedia platform for ethnomusicological research. + This platform, Telemeta, is designed to fit the professional requirements of both sound archivists, researchers and musicians, and to allow them to work together on large amounts of music data. The first prototype of this platform has been online since 2010 and is now fully operational. It has been used on a daily basis for ethnomusicological studies since 2011. -Recently, an open-source audio analysis framework, TimeSide, has been developed to bring automatic music analysis capabilities to the web platform and thus have turned Telemeta into a complete resource for Computational Ethnomusicology \cite{Tzanetakis_2007_JIMS, Gomez_JNMR_2013}. The Section~\ref{sec:TimeSide} focuses on this framework. +Recently, an open-source audio analysis framework, TimeSide, has been developed to bring automatic music analysis capabilities to the web platform; thus, Telemeta has become a complete resource for Computational Ethnomusicology \cite{Tzanetakis_2007_JIMS, Gomez_JNMR_2013}. Section~\ref{sec:TimeSide} focuses on this framework. -The benefits of this collaborative platform for humanities and social sciences research apply to numerous aspects of the field of ethnomusicology, ranging from musical analysis to comparative history and anthropology of music, as well as to the fields of anthropology, linguistics and acoustics. Some of these benefits have been mentionned in several ethnomusicological publications \cite{Simmonot_IASA_2011, Julien_IASA_2011, Simonnot_ICTM_2014}. -The current and potential applications of such a platform thus raises the needs and the underlying challenges of implementing an online digital tool to support contemporary scientific research. +This collaborative platform for humanities and social sciences research support numerous aspects of the field of ethnomusicology, ranging from musical analysis to comparative history and the anthropology of music. The platform also provides a useful resources for the fields of anthropology, linguistics and acoustics. Some of these benefits have been mentionned in several ethnomusicological publications \cite{Simmonot_IASA_2011, Julien_IASA_2011, Simonnot_ICTM_2014}. +Here, we discuss the features of this online platform. \section{The Telemeta platform}\label{sec:Telemeta} \subsection{Web audio content management features and architecture} -The primary purpose of the project is to provide the communities of researchers working on audio materials with a scalable system to access, preserve and share sound items along with associated metadata that contains key information on the context and significance of the recording. -Tele\-meta\footnote{\url{http://telemeta.org}}, as a free and open source software\footnote{Telemeta code is available under the CeCILL Free Software License Agreement}, is a unique scalable web audio platform for backuping, indexing, transcoding, analyzing, sharing and visualizing any digital audio or video file in accordance with open web standards. -The time-based nature of such audio-visual materials and some associated metadata as annotation raises issues of access and visualization at a large scale. Easy and on-demand access to these data, while listening to the recording, represents a significant improvement. +The primary purpose of the project is to provide researchers working on audio materials with a scalable system to access, preserve and share sound items along with associated metadata that contains key information on the context and significance of the recording. +Tele\-meta\footnote{\url{http://telemeta.org}}, is a free and open source\footnote{Telemeta code is available under the CeCILL Free Software License Agreement} web audio platform for indexing, transcoding, analyzing, sharing and visualizing any digital audio or video file in accordance with open web standards. +The temporal nature of such audio-visual materials and some associated metadata as annotation presents challenges for providing access and visualization at a large scale. Easy and on-demand access to these data, while listening to the recording, represents a significant improvement. An overview of the Telemeta's web interface is illustrated in Figure~\ref{fig:Telemeta}. \begin{figure*}[htb] \centering @@ -182,7 +183,7 @@ An overview of the Telemeta's web interface is illustrated in Figure~\ref{fig:Te \end{subfigure} \caption{Telemeta web interface} \end{figure*} -Its flexible and streaming safe architecture is represented in Figure~\ref{fig:TM_arch}. +The flexible and streaming safe architecture is represented in Figure~\ref{fig:TM_arch}. \begin{figure}[htb] \centering \includegraphics[width=0.9\linewidth]{img/TM_arch.pdf} @@ -191,50 +192,50 @@ Its flexible and streaming safe architecture is represented in Figure~\ref{fig:T \end{figure} The main features of Telemeta are: \begin{itemize} - \item Pure HTML5 web user interface including dynamic forms - \item On-the-fly audio analyzing, transcoding and metadata - embedding in various multimedia formats + \item Pure HTML5 web user interface including dynamic forms. + \item On-the-fly audio analyzing, transcoding and metadata. + embedding in various multimedia formats. \item Social editing with semantic ontologies, smart workflows, realtime tools, human or automatic annotations and - segmentations + segmentations. \item User management with individual desk, playlists, profiles - and group access rights - \item High level search engine (geolocation, instruments, ethnic groups, etc...) - \item Data providers : DublinCore, OAI-PMH, RSS, XML, JSON and other - \item Multi-language support (currently english and french) + and group access rights. + \item High level search engine (geolocation, instruments, ethnic groups, etc...). + \item Data providers : DublinCore, OAI-PMH, RSS, XML, JSON and other. + \item Multi-language support (currently english and french). \end{itemize} Beside database management, the audio support is mainly provided through an external component, TimeSide, which is described in Section~\ref{sec:TimeSide}. \subsection{Metadata}\label{sec:metadata} -In addition to the audio data, an efficient and dynamic management of the associated metadata is also necessary. Consulting metadata grants both an exhaustive access to valuable information about the source of the data and to the related work of peer researchers. +In addition to the audio data, an efficient and dynamic management of the associated metadata is also necessary. Metadata provides valuable information about the source of the data and to the related work of peer researchers. Dynamically handling metadata in a collaborative manner optimizes the continuous process of knowledge gathering and the enrichment of the materials in the database. One of the major challenges is the standardization of audio and metadata formats with the aim of long-term preservation and usage of the different materials. The compatibility with other systems is facilitated by the integration of the metadata standards protocols Dublin Core\footnote{{Dublin Core} Metadata Initiative, \url{http://dublincore.org/}} and \emph{OAI-PMH} (Open Archives Initiative Protocol for Metadata Harvesting)\footnote{\url{http://www.openarchives.org/pmh/}}. The metadata includes two different kinds of information about the audio item: contextual information and analytical information of the audio content. \subsubsection{Contextual Information} -In an ethnomusicological framework, contextual information may include details about the location where the recording has been made, the instruments, the population, the title of the musical piece, the cultural elements related to the musical item, the depositor, the collector, the year of the recording and the year of the publication. -Moreover, through the platform, diverse materials related to the archives can be stored, such as iconographies (digitalized pictures, scans of booklet and field notes, and so on), hyperlinks and biographical information about the collector. +In an ethnomusicological framework, contextual information may include details about the location where the recording was made, the instruments, the population, the title of the musical piece, the cultural elements related to the musical item, the depositor, the collector, the year of the recording and the year of publication of papers describing the work. +Moreover, through the platform, diverse materials related to the archives can be stored, such as iconographies (digitalized pictures, scans of booklets and field notes, and so on), hyperlinks and biographical information about the collector. \subsubsection{Descriptive and analytical information on the audio content} -The second type of metadata consists in information about the audio content itself. This metadata can relate to the global content of the audio item or provide temporally-indexed information. It should also be noted that such information can be produced either by a human expert or by an automatic computational audio analysis (see Section~\ref{sec:TimeSide} below). +The second type of metadata consists of information about the audio content itself. This metadata can provide information about the global content of the audio item or provide temporally-indexed information. It should also be noted that such information can be produced either by a human expert or by an automatic computational audio analysis (see Section~\ref{sec:TimeSide} below). \squeezeup\paragraph{Visual representation and segmentation} As illustrated in Figure~\ref{fig:sound_representation}, the TimeSide audio player embedded in the Telemeta web page view of a sound item allows for a selection of various visual representations of the sound (e.g. waveforms and spectrograms, see section~\ref{sec:TimeSide} for details) and some representations of computational analysis. -Among those automatic analysis some can produce a list of of time-segments associated with labels. -Those labels have been specified by the partners of the DIADEMS project (see Section~\ref{sec:Diadems} to be relevant for ethnomusicological studies (e.g. detection of spoken voice versus sang one, chorus, musical instrument categories, and so on). +Automatic analysis can produce a list of time-segments associated with labels. +Those labels have been specified by the partners of the DIADEMS project to be relevant for ethnomusicological studies (e.g. detection of spoken versus singing voices, chorus, musical instrument categories, and so on, see Section~\ref{sec:Diadems}). \squeezeup\paragraph{Annotations} -As illustrated in Figure~\ref{fig:Telemeta}, the embedded audio player also enables to annotate the audio content through time-coded markers. -Such annotations consist in a title and a free text field associated with a given time position. +As illustrated in Figure~\ref{fig:Telemeta}, the embedded audio player also enables annotation of the audio content through time-coded markers. +Such annotations consist of a title and a free text field associated with a given time position. -Ethnomusicologists, archivists, as well as anthropologists, linguists and acousticians working on sound documents can create their own annotations and share them with colleagues. These annotations are accessible from the sound archive item web page and are indexed through the database. +Ethnomusicologists, archivists, anthropologists, linguists and acousticians working on sound documents can create their own annotations and share them with colleagues. These annotations are accessible from the sound archive item web page and are indexed through the database. -It should be noted that the possibility for experts to annotate time-segments over a zoomable representation of the sound is currently under development in order to improve the accuracy and the quality of time-segment base annotations. +The possibility for experts to annotate time-segments over a zoomable representation of the sound is currently under development in order to improve the accuracy and the quality of time-segment-based annotations. \section{TimeSide, an audio analysis framework}\label{sec:TimeSide} -As illustrated in Figure~\ref{fig:TM_arch}, one specificity of the Telemeta architecture is to rely on an external component, TimeSide\footnote{\url{https://github.com/yomguy/TimeSide}}, that offers audio player web integration together with audio signal processing analysis capabilities. -TimeSide is an audio analysis and visualization framework based on both Python and JavaScript languages to provide state-of-the-art signal processing and machine learning algorithms together with web audio capabilities for display and streaming. +As illustrated in Figure~\ref{fig:TM_arch}, the Telemeta architecture relies on an external component, TimeSide\footnote{\url{https://github.com/yomguy/TimeSide}}, that offers audio player web integration together with audio signal processing analysis capabilities. +TimeSide is an audio analysis and visualization framework based on both Python and JavaScript languages that provides state-of-the-art signal processing and machine learning algorithms together with web audio capabilities for displaying and streaming files. Figure~\ref{fig:TimeSide_Archi} illustrates the overall architecture of TimeSide together with the data flow between TimeSide and the Telemeta web-server. \begin{figure}[htbp] \centering @@ -246,98 +247,96 @@ Figure~\ref{fig:TimeSide_Archi} illustrates the overall architecture of TimeSide TimeSide provides the following main features: \begin{itemize} \item Secure archiving, editing and publishing of audio files over - internet -\item Smart audio player with enhanced visualization (waveform, spectrogram) -\item Multi-format support: decodes the vast majority of audio and video formats through Gstreamer and transcodes them with smart streaming and caching methods% (FLAC, OGG, MP3, WAV and WebM) - % \item \emph{Playlist management} for all users with CSV data export -\item On-the-fly audio analyzing, transcoding and metadata - embedding based on an easy plugin architecture + the internet. +\item Smart audio player with enhanced visualization (waveform, spectrogram). +\item Multi-format support: decodes the vast majority of audio and video formats through Gstreamer and transcodes them with smart streaming and caching methods. +\item On-the-fly audio analysis, transcoding and metadata embedding based on an easy plugin architecture. \end{itemize} \subsection{Audio features extraction} -In order to implement Music Information Retrieval (MIR) analysis methods to be carried out over a large corpus for ethnomusicological studies, TimeSide incorporates some state-of-the-art audio feature extraction libraries such as Aubio\footnote{\url{http://aubio.org/}} \cite{brossierPhD}, Yaafe\footnote{\url{https://github.com/Yaafe/Yaafe}} \cite{yaafe_ISMIR2010} and Vamp plugins\footnote{ \url{http://www.vamp-plugins.org}}. -Given the extracted features, every sound item in a given collection can be automatically analyzed. The results of this analysis can be stored in a scientific file format (e.g. NumPy format and HDF5), exported to sound visualization and annotation softwares like sonic visualizer \cite{cannam2006sonic}, or serialized to the web browser through common markup languages: XML, JSON and YAML. +In order to implement Music Information Retrieval (MIR) analysis methods for large corpora for ethnomusicological studies, TimeSide incorporates some state-of-the-art audio feature extraction libraries such as Aubio\footnote{\url{http://aubio.org/}} \cite{brossierPhD}, Yaafe\footnote{\url{https://github.com/Yaafe/Yaafe}} \cite{yaafe_ISMIR2010} and Vamp plugins\footnote{ \url{http://www.vamp-plugins.org}}. +Given the extracted features, every sound item in a given collection can be automatically analyzed. The results of this analysis can be stored in a scientific file format (e.g. NumPy format and HDF5), exported to sound visualization and annotation software like sonic visualizer \cite{cannam2006sonic}, or serialized to the web browser through common markup languages: XML, JSON and YAML. -As an open-source framework and given its architecture and the flexibility provided by Python, the implementation of many audio and music analysis algorithms can be considered and then apply to audio archives through Telemeta. So, it is a very interesting platform for researchers in computational musicology to develop and evaluate their algorithms. +Telemeta's open-source framework and architecture, combined with the flexibility provided by Python, facilitates the implementation of many audio and music analysis algorithms and their application to audio archives. Thus, Telemeta is an interesting platform for researchers in computational musicology who wish to develop and evaluate their algorithms. \section{Sound archives of the \\CNRS - Musée de l'Homme}\label{sec:archives-CREM} -Since June 2011, the Telemeta platform is used by the Sound archives of the CNRS - Musée de l'Homme\footnote{\url{http://archives.crem-cnrs.fr}} and managed by the CREM. According to the CREM specific aims, the Telemeta platform makes these archives available for researchers, students and, when copyright allows it, to a broader audience. Through this platform, these archives can be shared, discussed and worked on. +Since June 2011, the Telemeta platform has been used by the Sound archives of the CNRS - Musée de l'Homme\footnote{\url{http://archives.crem-cnrs.fr}} and managed by the CREM. According to the CREM specific aims, the Telemeta platform makes these archives available for researchers, students and (when copyright allows) to a broader audience. Through this platform, these archives can be shared, discussed and analyzed. The Telemeta platform has also been deployed for the sound archives of the \emph{Laboratory of Musical Acoustic} of the ``Jean Le Rond d'Alembert Institute''\footnote{\url{http://telemeta.lam-ida.upmc.fr/}. Online since 2012, these archives consist in recordings of a wide range of musical instruments, mostly including solo recording of traditional instruments and illustrating various playing techniques and are used as materials for research in acoustics.}. \subsection{Archiving research materials} -The Sound archives of the CNRS - Musée de l'Homme is one of the most important in Europe and gather commercial as well as unpublished recordings of music and oral traditions from around the world, collected by researchers attached to numerous research institutions across the world, among which some prominent figures of the field of ethnomusicology (among which Brailoiu, Lomax, Shaeffner, Rouget and Elkin). +The Sound archives of the CNRS - Musée de l'Homme is one of the most important in Europe and contains commercial and unpublished recordings of music and oral traditions from around the world, collected by researchers attached to numerous research institutions across the world, including prominent figures of the field of ethnomusicology (among which Brailoiu, Lomax, Shaeffner, Rouget and Elkin). -The platform offers access to records collection (nearly 3700 hours, e.g. more than 5000 discs, many of which are very rare) and to 4000 hours of unpublished recordings, as early expeditions (e.g. Dakar-Djibouti (1932), Ogooué-Congo (1946)). Most of the recordings come from the fieldwork of researchers in all the continents. -More than 110 years of the world's oral culture are now available online, from the 1900 Universal Exhibition of Paris up to the recent digital recordings. The sharing of data allows several people to collaborate to the enrichment of the database. Today, 47,200 items are in the database, and more than 26,000 sound files have been included (12,000 sounds on free access since May 2014). Recently, the CREM has decided to give full access to the records published by the CNRS-Musée de l’Homme (Chant du Monde/Harmonia Mundi)\footnote{\url{http://archives.crem-cnrs.fr/archives/fonds/CNRSMH_Editions/}} which distribution stopped ten years ago. +The platform offers access to record collections (nearly 3700 hours, e.g. more than 5000 discs, many of which are very rare) and to 4000 hours of unpublished recordings, from early research expeditions (e.g. Dakar-Djibouti (1932), Ogooué-Congo (1946)). Most of the recordings come from the fieldwork of researchers in all continents. +More than 110 years of the world's oral culture are now available online, from the 1900 Universal Exhibition of Paris to recent digital recordings. The sharing of data allows several people to collaborate on the enrichment of the database. Today, 47,200 items are in the database, and more than 26,000 sound files have been included (12,000 sounds on free access since May 2014). Recently, the CREM has decided to give full access to the records published by the CNRS-Musée de l’Homme (Chant du Monde/Harmonia Mundi)\footnote{\url{http://archives.crem-cnrs.fr/archives/fonds/CNRSMH_Editions/}}, the distribution of which stopped ten years ago. As a web platform, this tool is also a way to cross borders, to get local populations involved in their own cultural heritage and to offer resources to researchers from all over the world. \subsection{Uses and users of digital sound archives} - Through the few years since the sound archive platform has been released, it appears to support three main activities: archive, research and education (academic or not). These usages are those of archivists, researchers (ethnomusicologists, anthropologists and linguists), students and professors of these disciplines. Nonetheless, a qualitative survey showed that other disciplines (such as art history) found some use of the platform to foster and/or deepen individual research. The unexpectedly broad uses of the sound archives once digitized and accessible emphasize the necessity and the benefits of such database. -From the standpoint of archive development, the long-term preservation of the archives is ensured while, thanks to the collaborative nature of the platform, users can cooperate to continuously enrich metadata associated with a sound document and submit their own archives to protect them. Furthermore, it allows fulfilling the ethical task of returning the recorded music to the communities who produced it. -Researchers from different institutions can work together on specific audio materials as well as conduct individual research in both synchronic and diachronic perspective, on their own material, material of others or both. -When use for education, the platform provides a wide array of teaching materials to illustrate the work of students as well as support teaching \emph{curricula}. +In the few years since the sound archive platform has been released, it has supported three main activities: archiving, research and education (both academic and non-academic). Primary users of the platform are archivists, researchers (ethnomusicologists, anthropologists and linguists), students and professors of these disciplines. Nonetheless, a qualitative survey showed that other disciplines (such as art history) have used the platform to foster and/or deepen individual research. The unexpectedly broad uses of the sound archives emphasize the necessity and the benefits of such database. +From the standpoint of archive development, the long-term preservation of the archives is ensured while, thanks to the collaborative nature of the platform, users can cooperate to continuously enrich metadata associated with a sound document and submit their own archives to protect them. Furthermore, it facilitates the ethical task of returning the recorded music to the communities who produced it. +Researchers from different institutions can work together on specific audio materials and conduct individual research from both synchronic and diachronic perspectives on their own material, the material of others, or both. +When used for education, the platform provides a wide array of teaching materials to illustrate the work of students as well as support teaching curricula. \section{Expanding development: the DIADEMS project}\label{sec:Diadems} -The goals and expectations of the platform are of many kinds and expand through time, as users experience new ways to work with the archives database and request new tools to broaden the scope of their research activities linked to it. The reflection collectively engaged by engineers and researchers on the use of the sound archives database led us to set up a large scale project called DIADEMS (\emph{Description, Indexation, Access to Ethnomusicological and Sound Documents})\footnote{\url{http://www.irit.fr/recherches/SAMOVA/DIADEMS/en/welcome/}}. +The many goals and expectations of the platform expand through time, as users experience new ways to work with the archives and request new tools to broaden the scope of their research activities linked to it. Comments from engineers and researchers on their usage of the sound archives database led us to set up a large scale project called DIADEMS (\emph{Description, Indexation, Access to Ethnomusicological and Sound Documents})\footnote{\url{http://www.irit.fr/recherches/SAMOVA/DIADEMS/en/welcome/}}. %DIADEMS is a French national research program, started in January 2013, with three IT research labs (IRIT\footnote{Institut de Recherche en Informatique de Toulouse}, , , LIMSI\footnote{Laboratoire d’Informatique pour la Mécanique et les Sciences de l’Ingénieur}, LABRI\footnote{Laboratoire Bordelais de Recherche en Informatique})\comment{TF: + LAM + labo ethno + Parisson. Plutôt dire a collaboration between ethno + IT} -Started in January 2013, the French national research program DIADEMS is a multi-disciplinary project whose consortium includes research laboratories from \emph{ Science and Technology of Information and Communication}\footnote{IRIT (Institute of research in computing science of Toulouse), LABRI (Bordeaux Computer Science Research Laboratory), LIMSI (Laboratory of computing and mechanics for engineering sciences), LAM (Laboratory of Musical Acoustic, Jean Le Rond d'Alembert Institute)} (IT) domain, \emph{Musicology and Ethnomusicology}\footnote{LESC (Laboratory of Ethnology and Comparative Sociology), MNHN (National Museum of Natural History)} domain and Parisson, a company involved in the development of Telemeta. +Started in January 2013, the French national research program DIADEMS is a multi-disciplinary project whose consortium includes research laboratories from the \emph{ Science and Technology of Information and Communication}\footnote{IRIT (Institute of research in computing science of Toulouse), LABRI (Bordeaux Computer Science Research Laboratory), LIMSI (Laboratory of computing and mechanics for engineering sciences), LAM (Laboratory of Musical Acoustic, Jean Le Rond d'Alembert Institute)} (IT) domain, the \emph{Musicology and Ethnomusicology}\footnote{LESC (Laboratory of Ethnology and Comparative Sociology), MNHN (National Museum of Natural History)} domain and Parisson, a company involved in the development of Telemeta. -The goal of the DIADEMS project is to develop computer tools to automatically index the recording content directly from the audio signal in order to improve the access to and the indexation of this vast ethnomusicological archive. Numerous ethnomusicological recordings contain speech and other types of sound that we categorized as sounds from the environment (such as rain, biological sounds, engine noise and so on) and sounds generated by the recording (such as sound produced by the wind on the microphone or sounds resulting from the defect of the recording medium). The innovation of this project is to automatize the indexation of the audio recordings directly from the recorded sound itself. Ongoing works consist in implementing advanced classification, indexation, segmentation and similarity analysis methods dedicated to ethnomusicological sound archives. Besides music analysis, such automatic tools also deal with speech and other types of sounds classification and segmentation to enable a more exhaustive annotation of the audio materials. +The goal of the DIADEMS project is to develop computer tools to automatically index the recording content directly from the audio signal in order to improve access to and indexation of this vast ethnomusicological archive. Numerous ethnomusicological recordings contain speech and other types of sound that we categorized as sounds from the environment (such as rain, biological sounds, engine noise and so on) and sounds generated by the recording process (such as sound produced by the wind on the microphone or sounds resulting from defects of the recording medium). The innovation of this project is to automatize the indexation of the audio recordings directly from the recorded sound itself. Ongoing work consists of implementing advanced classification, indexation, segmentation and similarity analysis methods dedicated to ethnomusicological sound archives. Besides music analysis, such automatic tools also deal with speech and other types of sounds classification and segmentation to enable a more exhaustive annotation of the audio materials. %The goal of Diadems project is to propose a set of tools for automatic analysis of audio documents which may contain fields recordings: speech, singing voice, instrumental music, technical noises, natural sounds, etc. The innovation is to automatize the indexation of audio recordings directly from the audio signal itself, in order to improve the access and indexation of anthropological archives. Ongoing works consist in implementing advanced classification, segmentation and similarity analysis methods, specially suitable to ethnomusicological sound archives. The aim is also to propose tools to analyse musical components and musical structure. -The automatic analysis of ethnomusicological sound archives is considered as a challenging task. -Field recordings generally contain more sound sources, noise, and recording artefacts than those obtained in studio conditions, so the automatic analysis of these recordings requires methods having a stronger robustness. -Preliminary implementations of speech detection models, and speaker diarization methods, based on \cite{barras2006multistage} have been integrated to TimeSide. -While these models are well suited to radio-news recordings, the current developpement tasks consist to adapt them to the particular case of ethnographic archives. +The automatic analysis of ethnomusicological sound archives is considered to be a challenging task. +Field recordings generally contain more sound sources, noise, and recording artefacts than those obtained in studio conditions, so the automatic analysis of these recordings requires robust methods. +Preliminary implementations of speech detection models and speaker diarization methods based on \cite{barras2006multistage} have been integrated to TimeSide. +While these models are well suited to radio-news recordings, the current development task is to adapt them to the particular case of ethnographic archives. -In the context of this project, researchers from Ethnomusicological, Speech Processing and MIR communities are working together to specify the tasks to be addressed by automatic analysis tools. +In the context of this project, researchers from the Ethnomusicological, Speech Processing and MIR communities are working together to specify the tasks to be addressed by automatic analysis tools. \subsection{The method of a new interdisciplinary research} -In this research program, groups from different backgrounds are working together to specify the automatic analysis tools: IT developers, humanities researchers (anthropologists, ethnomusicologists, ethnolinguists) and specialists in speech processing and MIR. The first challenge was to initiate a common interest and a mutual understanding. In this process, DIADEMS gave us the opportunity to improve our understanding on the link between the semantics and acoustics of voice production. As a prelimirary work we attempted to first define vocal categories with a particular interest for liminal oral productions. At the border between speech and song, utterances such as psalmody or recitation are at the center of an old debate in ethnomusicology\footnote{A colloquium on liminal utterances between speech and song will be organized by the International Council for Traditional Music (ICTM) in May 2015 and hosted by the Centre of research in Ethnomusicology (CREM). A round table will be dedicated to the presentation of the main results and findings of the ANR project DIADEMS}. Gathering specialists from various fields, DIADEMS project goes well beyond the usual disciplinary boundaries. Our aim, through the study of a large range of audio components (pitch range, syllabic flow, metric, polyphonic and so on) is to define and characterize the variability of vocal productions, keeping in mind the semantic aspects. By doing so, we wish to reduce the traditional gap in academic studies between sounds and semantics and to propose combined analytical tools for the study of vocal production\footnote{As an example, research will be conducted on the recognition of "icons of crying" +In this research program, groups from different backgrounds are working together to specify the automatic analysis tools. Participants include IT developers, humanities researchers (anthropologists, ethnomusicologists, ethnolinguists) and specialists in speech processing and MIR. The first challenge was to initiate a common interest and a mutual understanding. In this process, DIADEMS gave us the opportunity to improve our understanding on the link between the semantics and acoustics of voice production. +In preliminary work, we attempted to first define vocal categories with a particular interest for liminal oral productions. At the border between speech and song, utterances such as psalmody or recitation are at the center of an old debate in ethnomusicology\footnote{A colloquium on liminal utterances between speech and song will be organized by the International Council for Traditional Music (ICTM) in May 2015 and hosted by the Centre of research in Ethnomusicology (CREM). A round table will be dedicated to the presentation of the main results and findings of the ANR project DIADEMS}. Gathering specialists from various fields, the DIADEMS project goes well beyond the usual disciplinary boundaries. Our aim, through the study of a large range of audio components (pitch range, syllabic flow, metric, polyphonic and so on) is to define and characterize the variability of vocal productions, keeping in mind the semantic aspects. By doing so, we wish to reduce the traditional gap in academic studies between sounds and semantics and to develop combined analytical tools for the study of vocal production\footnote{As an example, research will be conducted on the recognition of "icons of crying" in lamented utterances. As defined by Urban in \cite{Urban88}, "icons of crying" include cry break, voice inhalation, creaky voice and falsetto vowels.}. -One of the goals of the DIADEMS project is also to provide useful tools for musical analysis such as detection of musical instrument families, analysis of musical content (tonal, metric and rhythmic features), musical similarities and structure (chorus localisation, musical pattern replication). +One of the goals of the DIADEMS project is also to provide useful tools for musical analysis such as the detection of musical instrument families, analysis of musical content (tonal, metric and rhythmic features), musical similarities and structure (chorus localisation, musical pattern replication). The study follows three steps : \begin{enumerate} \item The development of tools and selection of a representative corpus - for each tool + for each tool. \item The evaluation of the proposed automatic analysis, in addition to - the human-led evaluations carried on the corpus selected -\item The development of a visual interface with an ergonomic access and - import of the results in the database + the human-led evaluations carried out on the selected corpus. +\item The development of a visual interface for accessing and contributing to the database. \end{enumerate} \subsection{Automatic tools for assisting indexation and annotation of audio documents} -A first concern was to develop an automated annotation component that could differentiate spoken from sung voice and from instrumental music. Regarding that task, the state-of-the-art methods have usually been specifically designed to fit the needs of radio broadcast data (i.e. clear recordings produced in studios) and are not adapted to face the sonic diversity of ethnomusicological field recordings. For these, more refined detection tools were needed to pick up sound events such as overlapping speeches, speech over music, as well as instrumental music mixed with singing and/or spoken interventions. -Beyond the implementation of tools detecting the beginning and the end sound signatures of magnetic, mechanical and digital recorders as well as tape noises and silences, numerous algorithms allow for complex automated analysis for a wide range of combinations of vocal and instrumental sounds. +A first concern was to develop an automated annotation component that could differentiate spoken from sung voice and from instrumental music. State-of-the-art methods for this tasks have usually been designed to fit the needs of radio broadcast data (i.e. clear recordings produced in studios) and are not adapted to the sonic diversity of ethnomusicological field recordings. For these, more refined detection tools are needed to pick up sound events such as overlapping speech from multiple speakers, speech over music, and instrumental music mixed with singing and/or spoken interventions. +Beyond the implementation of tools for detecting the beginning and end of sound signatures of magnetic, mechanical and digital recorders as well as tape noises and silences, the platform provides algorithms for the complex automated analysis of a wide range of combinations of vocal and instrumental sounds. -\subsubsection{Analysis of recordings sessions} -A primary concern of people dealing with audio materials coming from field recordings is to quickly and automatically identify the position of the start and the stop time of a recording session. +\subsubsection{Analysis of recording sessions} +A primary concern of people dealing with field recordings is the fast and accurate identification of the start and the stop times of a recording session. -In the case of digital recorder, the proposed system for localizing the start of recording sessions is based on the observation that on such recorders, the powering of the recorder implies a characteristic and quite reproducible perturbation of the signal as illustrated in Figure~\ref{plop}. Therefore, this perturbation can be modeled by using some reference models of the shape of the temporal energy. On the analyzed recordings, the evolution of the temporal energy of low energy segments is then compared to the models by an euclidean distance using the best alignment of the energy shape. +In the case of digital recorders, the proposed system for localizing the start of recording sessions is based on the observation that on such recorders, the powering of the recorder implies a characteristic and quite reproducible perturbation of the signal as illustrated in Figure~\ref{plop}. Therefore, this perturbation can be modeled by using some reference models of the shape of the temporal energy. On the analyzed recordings, the evolution of the temporal energy of low energy segments is then compared to the models by a euclidean distance using the best alignment of the energy shape. \begin{figure}[htb] \centering \includegraphics[width=0.8\linewidth]{img/plop.png} - \caption{Typical perturbation produced by the recorder powering, used to recognize session start} + \caption{Typical perturbation produced by the recorder powering-on, used to recognize the start time of a session.} \label{plop} \end{figure} \subsubsection{Analysis of speech and singing voice segments} -Quick identification and localization of spoken sections, particularly in rather long recordings, are relevant for all the disciplines involved in the project. The difficulties inherent in the sound materials led to the development of tools to automatically detect the occurrences of speech when performed simultaneously or alternatively with music; when numerous speakers interact and/or overlap with each others, with or without additional music or noises; and when voices modulate from speech to song, using a wide range of vocal techniques (recitation, narration, psalmody, back channel, and so on). The developed algorithms allow the analysis of the syllabic flow and the prosody of the speech. +The quick identification and localization of spoken sections, particularly in rather long recordings, is important for all the disciplines involved in this project. Difficulties inherent in the processing of sound materials led to the development of tools to automatically detect occurrences of speech when performed simultaneously with music. The detection of speech is important in a number of scenarios, such as when numerous speakers interact and/or overlap with each others---with or without additional music or noises---and when voices modulate from speech to song, using a wide range of vocal techniques (recitation, narration, psalmody, back channel, and so on). The developed algorithms allow the analysis of the syllabic flow and the prosody of the speech. Figure~\ref{fig:speech_detection} shows a visual example of how speech segmentation is rendered. %Source : CNRSMH_I_2013_201_001_01/ @@ -348,7 +347,7 @@ Figure~\ref{fig:speech_detection} shows a visual example of how speech segmentat \label{fig:speech_detection} \end{figure} -\squeezeup\paragraph{Speech segmentation, with 2 features: 4 Hz modulation energy and entropy modulation} +\squeezeup\paragraph{Speech segmentation with 2 features: 4 Hz modulation energy and entropy modulation} Speech signal has a characteristic energy modulation peak around the 4 Hertz syllabic rate \cite{Houtgast1985}. In order to model this property, the signal is filtered with a FIR band pass filter, centered on 4 Hertz. Entropy modulation is dedicated to discriminate between speech and music~\cite{Pinquier2003}. We first evaluate the signal entropy ($H=-\sum_{i=1}^{k}p_i\cdot \log_2(p_i)$, where $p_i$ denotes the probability of event~$i$). Entropy modulation values are usually larger for speech than for music. This measure is used to compute the entropy modulation on each segment. @@ -356,19 +355,20 @@ Entropy modulation is dedicated to discriminate between speech and music~\cite{P Speech activity detection is a prerequisite for several speech-related tasks to be integrated in the platform such as speech segmentation, speaker diarization and so on. For this approach, a Gaussian Mixture Model system operating at the frame level, is used to learn speech spectral properties from annotated data. MFCC, together with their first and second time derivative, log energy and zero crossing rate audio features are used to train the GMM models to discriminate speech frames from non speech frames. -ETAPE corpus \cite{gravier2012etape} (speech-based TV archives in French), and data obtained from the French Center of Research and Teaching on Amerindian Ethnology (carnival rituals in Mayan language) were used to train two distinct sets of models. -Both models have been serialized and integrated to TimeSide together with the GMM-based speech activity detection system, allowing the final user to choose a given model according to the properties of the media to analyze. +The ETAPE corpus \cite{gravier2012etape} (speech-based TV archives in French), and data obtained from the French Center of Research and Teaching on Amerindian Ethnology (carnival rituals in Mayan language) were used to train two distinct sets of models. +Both models have been serialized and integrated into TimeSide together with the GMM-based speech activity detection system, allowing the final user to choose a given model according to the properties of the media to analyze. \subsubsection{Analysis of music segments} -The DIADEMS project aims to provide useful tools for musical analysis in both research and teaching frameworks. So, the detection of segments of instrumental music along with the recognition of the different musical instrument categories is needed. Pushing the detection further into details, the implemented tools provide musicological information to support sound analysis (such as tonal, metric and rhythmic features) and allow detecting of similarities in melody, harmony and rhythm as well as musical pattern replications. +The DIADEMS project aims to provide useful tools for musical analysis in both research and teaching frameworks. Thus, the detection of segments of instrumental music along with the recognition of the different musical instrument categories is needed. The implemented tools provide musicological information to support sound analysis (such as tonal, metric and rhythmic features) and to allow the detection of similarities in melody, harmony and rhythm, as well as musical pattern replications. -\squeezeup\paragraph{Music segmentation, with 2 features based on a segmentation algorithm} -This segmentation is provided by the Forward-Backward Divergence algorithm, which is based on a statistical study of the acoustic signal \cite{Obrecht1988}. The speech signal is assumed to be composed of a sequence of quasi-stationary units that can be seen as alternate periods of transient and steady parts (steady parts are mainly vowels). We characterize each of these units by an Auto Regressive (AR) Gaussian model. The method consists in performing a detection of changes in AR models. Indeed, music is usually much more constant than speech, that is to say the number of changes (segments) will be smaller for music than for speech. To estimate this, we count the number of segments per second of signal. The number of segments is the first discriminative feature for music segmentation. +\squeezeup\paragraph{Music segmentation with 2 features based on a segmentation algorithm} +This segmentation is provided by the Forward-Backward Divergence algorithm, which is based on a statistical study of the acoustic signal \cite{Obrecht1988}. The speech signal is assumed to be composed of a sequence of quasi-stationary units that can be seen as alternate periods of transient and steady parts (steady parts are mainly vowels). We characterize each of these units by an Auto Regressive (AR) Gaussian model. The method detects changes in AR models. Indeed, music is usually much more constant than speech, that is to say the number of changes (segments) will be smaller for music than for speech. To estimate this, we count the number of segments per second of signal. The number of segments is the first discriminative feature for music segmentation. The segmentation algorithm generally produces longer segments for music than for speech. We chose to model the segment duration by an inverse Gaussian distribution (or Wald distribution) which is indeed the second feature providing music segmentation. \squeezeup\paragraph{Monophony / Polyphony segmentation} -A "monophonic" sound is defined as one note played at a time (either played by an instrument or sung by a singer), while a "polyphonic" sound is defined as several notes played simultaneously. The parameters extracted from the signal come from the YIN algorithm, a well known pitch estimator \cite{DeCheveigne2002}. Besides F0, this estimator provides an additional numerical value that can be interpreted as the inverse of a confidence indicator (the lower the value, the more reliable the estimated pitch). Considering that when there is a single note the estimated pitch is fairly reliable, and that when there are several simultaneous notes, the estimated pitch is not reliable, we take the short term mean and variance of this "confidence indicator" as parameters for the monophony / polyphony segmentation. The bivariate distribution of these two parameters is modelled using Weibull bivariate distributions \cite{Lachambre2011}. +A ``monophonic'' sound is defined as one note played at a time (either played by an instrument or sung by a singer), while a ``polyphonic'' sound is defined as several notes played simultaneously. The parameters extracted from the signal come from the YIN algorithm, a well known pitch estimator \cite{DeCheveigne2002}. Besides F0, this estimator provides an additional numerical value that can be interpreted as the inverse of a confidence indicator (the lower the value, the more reliable the estimated pitch). +Considering that the estimated pitch is fairly accurate when there is a single note, and that the estimated pitch is not reliable when there are several simultaneous notes, we take the short term mean and variance of this ``confidence indicator'' as parameters for the monophony / polyphony segmentation. The bivariate distribution of these two parameters is modelled using Weibull bivariate distributions \cite{Lachambre2011}. An example of the segmentation produced by this method is illustrated in Figure~\ref{fig:Monopoly} % Source : CNRSMH_I_2000_008_001_04 \begin{figure}[htb] @@ -380,9 +380,9 @@ An example of the segmentation produced by this method is illustrated in Figure~ \end{figure} \squeezeup\paragraph{Automatic instrument classification} -For the detection of musical instrument, we choose to follow the Hornbostel–Sachs system of musical instrument classification as first published in \cite{taxonomy_sachs2} and later translated in \cite{taxonomy_sachs}. This system is the most widely used system for classifying musical instruments by ethnomusicologists and organologists. It was extended by in more recent systems like the one proposed by Geneviève Dournon in \cite{Dournon92} and the \emph{RAMEAU} reference from the French national library (BnF)\footnote{\url{http://catalogue.bnf.fr/ark:/12148/cb119367821/PUBLIC}}. +For the detection of musical instruments, we choose to follow the Hornbostel–Sachs system of musical instrument classification as first published in \cite{taxonomy_sachs2} and later translated in \cite{taxonomy_sachs}. This system is the most widely used system for classifying musical instruments by ethnomusicologists and organologists. It was extended by in more recent systems like the one proposed by Geneviève Dournon in \cite{Dournon92} and the \emph{RAMEAU} reference from the French national library (BnF)\footnote{\url{http://catalogue.bnf.fr/ark:/12148/cb119367821/PUBLIC}}. -We choose to develop tools to detect the four musical instrument families (cordophones, aerophones, idiophones, membranophones), but also refine subdivisions related to the playing techniques, identifying whether each instrument is blowed, bowed, plucked, struck or clincked by focusing on the 8 classes of instrument shown in Figure~\ref{fig:instruments}. +We choose to develop tools to detect the four musical instrument families (cordophones, aerophones, idiophones, membranophones), but also to refine subdivisions related to the playing techniques, identifying whether each instrument is blown, bowed, plucked, struck or clincked by focusing on the 8 classes of instrument shown in Figure~\ref{fig:instruments}. \begin{figure}[htb] \centering @@ -407,13 +407,13 @@ The Linear Discriminant Analysis (LDA) method~\cite{lda_book} is used to compute or linear combination of all descriptors which maximizes the average distance between classes while minimizing the distance between individuals of the same class. -Each class is simply modeled into the classification space by its centroid (mean vector computed over the selected descriptors of the individuals which compose the class). -Thus, the classification task consists in projecting each tested sound described by a vector of descriptors -to the classification space and to select the closer class (in term of Euclidean distance to the class centroid). +Each class is simply modeled in the classification space by its centroid (mean vector computed over the selected descriptors of the individuals which compose the class). +Thus, the classification task consists of projecting each tested sound described by a vector of descriptors +into the classification space and selecting the closer class (in term of Euclidean distance to the class centroid). -This simple and computationally efficient method obtains about 75\% of accuracy -using the 20 most relevant descriptors projected on a 8-dimensions discriminative space. In a more detailed study~\cite{ismir14_dfourer}, -this promising result was shown comparable with state-of-the art methods applied for the classification of western instruments recorded in studio condition. +This simple and computationally efficient method reaches an accuracy of 75\% +using the 20 most relevant descriptors projected on an 8-dimensional discriminative space. In a more detailed study~\cite{ismir14_dfourer}, +this promising result was shown to be comparable with state-of-the art methods applied to the classification of western instruments recorded in studio conditions. % \begin{figure}[!ht] % \centering @@ -423,26 +423,25 @@ this promising result was shown comparable with state-of-the art methods applied % \end{figure} -\subsection{Evaluation and sought improvements} +\subsection{Evaluation and future improvements} -At the end of the first step of the project, interesting preliminary results have been obtained regarding start of recording session detection, speech recognition, singing voice recognition and musical instrument family classification. +At the end of the first step of the project, interesting preliminary results have been obtained regarding the detection of start times of recording sessions, speech recognition, singing voice recognition and musical instrument family classification. -Through a collaborative work, ethnomusicologists, ethnolinguists and engineers are currently evaluating, correcting and refining the implemented tools, with the expectation that this work will lead to positive results, so these new tools can be integrated into the Telemeta platform. +Through collaborative work, ethnomusicologists, ethnolinguists and engineers are currently evaluating, correcting and refining the implemented tools, with the expectation that these new tools will be integrated into the Telemeta platform. -The robustness of all these processing are assessed using criteria defined by the final users: teachers, students, researchers or musicians. Annotation tools, as well as the provided annotations, will be integrated in the digitalized database. +The robustness of these processing is assessed using criteria defined by the final users: teachers, students, researchers and musicians. Annotation tools, as well as the provided annotations, will be integrated in the digitalized database. -Further work on the user interface aims to enhance the visualization experience with time and frequency zooming capabilities, in the hope that it will improve the accuracy and the quality of time-segment based annotations. One of the remaining issues is to develop tools to generate results in line with the server processor and according to the capabilities of Internet browsers while managing the workflow. +Further work on the user interface will enhance the visualization experience with time and frequency zooming capabilities, in the hope that it will improve the accuracy and the quality of time-segment based annotation. One of the remaining goals is to develop tools to generate results online and to make use of the capabilities of Internet browsers while managing the workflow. \section{Conclusion} - The Telemeta open-source framework brings researchers in humanities and social sciences a new platform to efficiently distribute, share and work on their research on musical and sound materials. -This platform offers automatic music analysis capabilities through the external component, TimeSide that provides a flexible computational analysis engine together with web serialization and visualization options. -It brings an appropriate processing framework for researchers in computational ethnomusicology to develop and evaluate their algorithms. + The Telemeta open-source framework provides a new platform for researchers in humanities and social sciences to efficiently distribute, share and work on their research on musical and sound materials. +This platform offers automatic music analysis capabilities through the external component, TimeSide, which provides a flexible computational analysis engine together with web serialization and visualization options. +The Telemeta platform provides an appropriate processing framework for researchers in computational ethnomusicology to develop and evaluate their algorithms. Deployed to manage the CNRS - Musée de l’Homme sound archives, the Telemeta platform has been conceived and adapted to generate tools in line with the needs of users. Thanks to the collaborative nature of the platform, users can continuously enrich metadata associated with sound archives. - -The benefits of this collaborative platform for the field of ethnomusicology apply to numerous aspects of research, ranging from musical analysis in a diachronic and synchronic comparative perspective, as well as the long-term preservation of sound archives and the support of teaching materials for education. +The benefits of this collaborative platform for the field of ethnomusicology apply to numerous aspects of research, ranging from musical analysis in diachronic and synchronic comparative perspectives, as well as the long-term preservation of sound archives and the support of teaching materials for education. \section{Acknowledgments} -- 2.39.5