Add IRIT Speech detectors

diff --git a/timeside/analyzer/__init__.py b/timeside/analyzer/__init__.py
index 4c1c6804ca0f9479010cbe5f2abcbf164f195f0f..ab43ac77000bb035717adc14180dc378b40d1001 100644 (file)
--- a/timeside/analyzer/__init__.py
+++ b/timeside/analyzer/__init__.py
@@ -7,3 +7,5 @@ from aubio_pitch import *
 from aubio_mfcc import *
 from aubio_melenergy import *
 from aubio_specdesc import *
+from irit_speech_entropy import *
+from irit_speech_4hz import *

diff --git a/timeside/analyzer/utils.py b/timeside/analyzer/utils.py
index b7c8c610fda5e3c5ab95f2383f3a8bfd4130d1eb..43b52e5d461a59b4132653a404f7dd23843ec332 100644 (file)
--- a/timeside/analyzer/utils.py
+++ b/timeside/analyzer/utils.py
@@ -34,4 +34,184 @@ def downsample_blocking(frames, hop_s, dtype='float32'):
     # blocking
     return downsampled.reshape(downsampled.shape[0] / hop_s, hop_s)
 
+def computeModulation(serie,wLen,withLog=True):
+        '''
+        Compute the modulation of a parameter centered. Extremums are set to zero.
+        
+        Args :
+            - serie       : list or numpy array containing the serie.
+            - wLen        : Length of the analyzis window in samples.
+            - withLog     : Whether compute the var() or log(var()) .    
+        
+        Returns :
+            - modul       : Modulation of the serie.
+        
+        '''
+        
+        modul = numpy.zeros((1,len(serie)))[0];
+        w = int(wLen/2)
+        
+        for i in range(w,len(serie)-w):
+            
+            d = serie[i-w:i+w]
+            if withLog:
+                d = numpy.log(d)
+            modul[i] = numpy.var(d)
+        
+        modul[:w] = modul[w]
+        
+        modul[-w:] = modul[-w-1]
+    
+        return modul;
+
+def segmentFromValues(values,offset=0):
+    '''
+    
+    '''
+    
+    seg = [offset,-1,values[0]]
+    segList = []
+    for i,v in enumerate(values) :
+
+        if not (v == seg[2]) :
+            seg[1] = i+offset
+            segList.append(tuple(seg))
+            seg = [i+offset,-1,v]
+            
+    seg[1] = i+offset
+    segList.append(tuple(seg))
+
+    return segList
+
+
+# Attention 
+# ---------
+#
+# Double emploi avec le calcul mfcc d'aubio. Voir pour la fusion...
+#                                              Maxime
+
+def melFilterBank(nbFilters,fftLen,sr) :
+    '''
+    Grenerate a Mel Filter-Bank
+        
+    Args :
+        - nbFilters  : Number of filters.
+        - fftLen     : Length of the frequency range.
+        - sr         : Sampling rate of the signal to filter. 
+    Returns :
+        - filterbank : fftLen x nbFilters matrix containing one filter by column.
+                        The filter bank can be applied by matrix multiplication 
+                        (Use numpy *dot* function).      
+    '''
+        
+    fh = float(sr)/2.0    
+    mh = 2595*numpy.log10(1+fh/700)
+        
+    step = mh/nbFilters;
+        
+    mcenter = numpy.arange(step,mh,step)
+        
+    fcenter = 700*(10**(mcenter/2595)-1)
+            
+    filterbank = numpy.zeros((fftLen,nbFilters));    
+        
+    for i,_ in enumerate(fcenter) :
+            
+        if i == 0 :
+            fmin = 0.0
+        else :
+            fmin = fcenter[i-1]
+                
+        if i == len(fcenter)-1 :
+            fmax = fh
+        else :
+            fmax = fcenter[i+1]    
+            
+        imin = numpy.ceil(fmin/fh*fftLen)
+        imax = numpy.ceil(fmax/fh*fftLen)
+            
+        filterbank[imin:imax,i] = triangle(imax-imin)
+    
+    return filterbank
+
+
+def triangle(length):
+    '''
+    Generate a triangle filter.
+        
+    Args :
+         - length  : length of the filter.
+    returns :
+        - triangle : triangle filter.    
+            
+    '''
+    triangle = numpy.zeros((1,length))[0]
+    climax= numpy.ceil(length/2)
+        
+    triangle[0:climax] = numpy.linspace(0,1,climax)
+    triangle[climax:length] = numpy.linspace(1,0,length-climax)
+    return triangle
+    
+
+def entropy(serie,nbins=10,base=numpy.exp(1),approach='unbiased'):
+        '''
+        Compute entropy of a serie using the histogram method.
+        
+        Args :
+            - serie     : Serie on witch compute the entropy
+            - nbins     : Number of bins of the histogram
+            - base      : Base used for normalisation
+            - approach  : String in the following set : {unbiased,mmse}
+                          for un-biasing value.
+    
+        Returns :
+            - estimate  : Entropy value
+            - nbias     : N-bias of the estimate
+            - sigma     : Estimated standard error
+            
+        Raises : 
+            A warning in case of unknown 'approach' value. 
+            No un-biasing is then performed 
+        
+        '''
+               
+        estimate = 0
+        sigma = 0
+        bins,edges = numpy.histogram(serie,nbins);
+        ncell = len(bins)
+        norm = (numpy.max(edges)-numpy.min(edges))/len(bins)
+    
+    
+        for b in bins :
+            if b == 0 :
+                logf = 0
+            else :
+                logf = numpy.log(b)
+            estimate = estimate - b*logf
+            sigma = sigma + b * logf**2
+            
+        count = numpy.sum(bins)
+        estimate=estimate/count;
+        sigma=numpy.sqrt( (sigma/count-estimate**2)/float(count-1) );
+        estimate=estimate+numpy.log(count)+numpy.log(norm);
+        nbias=-(ncell-1)/(2*count);
+    
+        if approach =='unbiased' :
+            estimate=estimate-nbias;
+            nbias=0;
+            
+        elif approach =='mmse' :
+            estimate=estimate-nbias;
+            nbias=0;
+            lambda_value=estimate^2/(estimate^2+sigma^2);
+            nbias   =(1-lambda_value)*estimate;
+            estimate=lambda_value*estimate;
+            sigma   =lambda_value*sigma;
+        else :
+            return 0
+                
+        estimate=estimate/numpy.log(base);
+        nbias   =nbias   /numpy.log(base);
+        sigma   =sigma   /numpy.log(base);
+        return estimate