o [i* @sddlmZddlZddlmZddlmZddddd d ddd d d ddf ddZdddd d ddd ddf ddZ  d%ddZ dddd d ddd ddf ddZ ddZ ddZ d&dd Z d'd!d"Zd#d$ZdS)()divisionN)sigproc)dct>皙?{Gz?  ףp= ?TcC t|fSNnumpyonesxrY/home/ubuntu/SoloSpeech/.venv/lib/python3.10/site-packages/python_speech_features/base.py  rc  Cspt||||||||| | \} }t| } t| ddddddd|f} t| | } | r6t|| dddf<| S)aXCompute MFCC features from an audio signal. :param signal: the audio signal from which to compute features. Should be an N*1 array :param samplerate: the samplerate of the signal we are working with. :param winlen: the length of the analysis window in seconds. Default is 0.025s (25 milliseconds) :param winstep: the step between successive windows in seconds. Default is 0.01s (10 milliseconds) :param numcep: the number of cepstrum to return, default 13 :param nfilt: the number of filters in the filterbank, default 26. :param nfft: the FFT size. Default is 512. :param lowfreq: lowest band edge of mel filters. In Hz, default is 0. :param highfreq: highest band edge of mel filters. In Hz, default is samplerate/2 :param preemph: apply preemphasis filter with preemph as coefficient. 0 is no filter. Default is 0.97. :param ceplifter: apply a lifter to final cepstral coefficients. 0 is no lifter. Default is 22. :param appendEnergy: if this is true, the zeroth cepstral coefficient is replaced with the log of the total frame energy. :param winfunc: the analysis window to apply to each frame. By default no window is applied. You can use numpy window functions here e.g. winfunc=numpy.hamming :returns: A numpy array of size (NUMFRAMES by numcep) containing features. Each row holds 1 feature vector. ortho)typeaxisnormNr)fbankrlogrlifter)signal sampleratewinlenwinstepnumcepnfiltnfftlowfreqhighfreqpreemph ceplifter appendEnergywinfuncfeatenergyrrrmfccs    r0cCr rrrrrrr%rc Cs|p|d}t||}t|||||| } t| |} t| d} t| dkttj | } t |||||} t | | j }t|dkttj |}|| fS)aCompute Mel-filterbank energy features from an audio signal. :param signal: the audio signal from which to compute features. Should be an N*1 array :param samplerate: the samplerate of the signal we are working with. :param winlen: the length of the analysis window in seconds. Default is 0.025s (25 milliseconds) :param winstep: the step between successive windows in seconds. Default is 0.01s (10 milliseconds) :param nfilt: the number of filters in the filterbank, default 26. :param nfft: the FFT size. Default is 512. :param lowfreq: lowest band edge of mel filters. In Hz, default is 0. :param highfreq: highest band edge of mel filters. In Hz, default is samplerate/2 :param preemph: apply preemphasis filter with preemph as coefficient. 0 is no filter. Default is 0.97. :param winfunc: the analysis window to apply to each frame. By default no window is applied. You can use numpy window functions here e.g. winfunc=numpy.hamming :returns: 2 values. The first is a numpy array of size (NUMFRAMES by nfilt) containing features. Each row holds 1 feature vector. The second return value is the energy in each frame (total energy, unwindowed) rrr) r preemphasisframesigpowspecrsumwherefinfofloatepsget_filterbanksdotT)r!r"r#r$r&r'r(r)r*r-framespspecr/fbr.rrrr#s    rc Cs&t||||||||| \} } t| S)aCompute log Mel-filterbank energy features from an audio signal. :param signal: the audio signal from which to compute features. Should be an N*1 array :param samplerate: the samplerate of the signal we are working with. :param winlen: the length of the analysis window in seconds. Default is 0.025s (25 milliseconds) :param winstep: the step between successive windows in seconds. Default is 0.01s (10 milliseconds) :param nfilt: the number of filters in the filterbank, default 26. :param nfft: the FFT size. Default is 512. :param lowfreq: lowest band edge of mel filters. In Hz, default is 0. :param highfreq: highest band edge of mel filters. In Hz, default is samplerate/2 :param preemph: apply preemphasis filter with preemph as coefficient. 0 is no filter. Default is 0.97. :returns: A numpy array of size (NUMFRAMES by nfilt) containing features. Each row holds 1 feature vector. )rrr) r!r"r#r$r&r'r(r)r*r.r/rrrlogfbankBs r?cCr rrrrrrrVrc  Cs|p|d}t||}t|||||| } t| |} t| dkttj| } t |||||} t | | j } t t d|dt| dt| ddf}t | || j | S)aJCompute Spectral Subband Centroid features from an audio signal. :param signal: the audio signal from which to compute features. Should be an N*1 array :param samplerate: the samplerate of the signal we are working with. :param winlen: the length of the analysis window in seconds. Default is 0.025s (25 milliseconds) :param winstep: the step between successive windows in seconds. Default is 0.01s (10 milliseconds) :param nfilt: the number of filters in the filterbank, default 26. :param nfft: the FFT size. Default is 512. :param lowfreq: lowest band edge of mel filters. In Hz, default is 0. :param highfreq: highest band edge of mel filters. In Hz, default is samplerate/2 :param preemph: apply preemphasis filter with preemph as coefficient. 0 is no filter. Default is 0.97. :param winfunc: the analysis window to apply to each frame. By default no window is applied. You can use numpy window functions here e.g. winfunc=numpy.hamming :returns: A numpy array of size (NUMFRAMES by nfilt) containing features. Each row holds 1 feature vector. rrr)rr1r2r3rr5r6r7r8r9r:r;tilelinspacesize)r!r"r#r$r&r'r(r)r*r-r<r=r>r.RrrrsscTs   .rDcCsdtd|dS)zConvert a value in Hertz to Mels :param hz: a value in Hz. This can also be a numpy array, conversion proceeds element-wise. :returns: a value in Mels. If an array was passed in, an identical sized array is returned. i# rg@)rlog10)hzrrrhz2melqsrGcCsdd|ddS)zConvert a value in Mels to Hertz :param mel: a value in Mels. This can also be a numpy array, conversion proceeds element-wise. :returns: a value in Hertz. If an array was passed in, an identical sized array is returned. i gF@rr)melrrrmel2hzysrJc Cs*|p|d}||dksJdt|}t|}t|||d}t|dt||}t||ddg} td|D]T} tt|| t|| dD]} | || || d|| | | | f<qOtt|| dt|| dD]} || d| || d|| d| | | f<qwq>| S)axCompute a Mel-filterbank. The filters are stored in the rows, the columns correspond to fft bins. The filters are returned as an array of size nfilt * (nfft/2 + 1) :param nfilt: the number of filters in the filterbank, default 20. :param nfft: the FFT size. Default is 512. :param samplerate: the samplerate of the signal we are working with. Affects mel spacing. :param lowfreq: lowest band edge of mel filters, default 0 Hz :param highfreq: highest band edge of mel filters, default samplerate/2 :returns: A numpy array of size nfilt * (nfft/2 + 1) containing filterbank. Each row holds 1 filter. rz%highfreq is greater than samplerate/2rr)rGrrAfloorrJzerosrangeint) r&r'r"r(r)lowmelhighmel melpointsbinrjirrrr9s "*&2r9cCsL|dkr$t|\}}t|}d|dttj||}||S|S)a8Apply a cepstral lifter the the matrix of cepstra. This has the effect of increasing the magnitude of the high frequency DCT coeffs. :param cepstra: the matrix of mel-cepstra, will be numframes * numcep in size. :param L: the liftering coefficient to use. Default is 22. L <= 0 disables lifter. rrg@)rshapearangesinpi)cepstraLnframesncoeffnliftrrrr s   r c Cs|dkrtdt|}dtddtd|dD}t|}tj|||fdfdd}t|D]}tt| |d|||d|d|||<q1|S) aCompute delta features from a feature vector sequence. :param feat: A numpy array of size (NUMFRAMES by number of features) containing features. Each row holds 1 feature vector. :param N: For each frame, calculate delta features based on preceding and following N frames :returns: A numpy array of size (NUMFRAMES by number of features) containing delta features. Each row holds 1 delta feature vector. rzN must be an integer >= 1rcSsg|]}|dqS)rr).0rUrrr szdelta..)rredge)mode) ValueErrorlenr4rNr empty_likepadr:rW)r.N NUMFRAMES denominator delta_featpaddedtrrrdeltas   8rn)rrrr r rNr )rKr rrN)r ) __future__rrpython_speech_featuresr scipy.fftpackrr0rr?rDrGrJr9r rnrrrrs0