+ jTARt^RIHt^RIt^RIHtHtHtH t ^RI H t ^RI H t ^RIHtHtHt]P$!]4t!RR] 4tR.tR#) z%Feature extractor class for SpeechT5.)AnyN)mel_filter_bankoptimal_fft_length spectrogramwindow_function)SequenceFeatureExtractor) BatchFeature)PaddingStrategy TensorTypeloggingcaa]tRt^toRtRR.tRV3RlV3Rlllt]RV3RlRll4tV3RlR lt RV3R lR llt RV3R lR llt V3RlV3Rllt Rt VtV;t#)SpeechT5FeatureExtractoraC Constructs a SpeechT5 feature extractor. This class can pre-process a raw speech signal by (optionally) normalizing to zero-mean unit-variance, for use by the SpeechT5 speech encoder prenet. This class can also extract log-mel filter bank features from raw speech, for use by the SpeechT5 speech decoder prenet. This feature extractor inherits from [`~feature_extraction_sequence_utils.SequenceFeatureExtractor`] which contains most of the main methods. Users should refer to this superclass for more information regarding those methods. Args: feature_size (`int`, *optional*, defaults to 1): The feature dimension of the extracted features. sampling_rate (`int`, *optional*, defaults to 16000): The sampling rate at which the audio files should be digitalized expressed in hertz (Hz). padding_value (`float`, *optional*, defaults to 0.0): The value that is used to fill the padding values. do_normalize (`bool`, *optional*, defaults to `False`): Whether or not to zero-mean unit-variance normalize the input. Normalizing can help to significantly improve the performance for some models. num_mel_bins (`int`, *optional*, defaults to 80): The number of mel-frequency bins in the extracted spectrogram features. hop_length (`int`, *optional*, defaults to 16): Number of ms between windows. Otherwise referred to as "shift" in many papers. win_length (`int`, *optional*, defaults to 64): Number of ms per window. win_function (`str`, *optional*, defaults to `"hann_window"`): Name for the window function used for windowing, must be accessible via `torch.{win_function}` fmin (`float`, *optional*, defaults to 80): Minimum mel frequency in Hz. fmax (`float`, *optional*, defaults to 7600): Maximum mel frequency in Hz. mel_floor (`float`, *optional*, defaults to 1e-10): Minimum value of mel frequency banks.. return_attention_mask (`bool`, *optional*, defaults to `True`): Whether or not [`~SpeechT5FeatureExtractor.__call__`] should return `attention_mask`. input_valuesattention_maskcb<V^8dQhRS[RS[RS[RS[RS[RS[RS[RS[R S[R S[R S[R S[/ #)  feature_size sampling_rate padding_value do_normalize num_mel_bins hop_length win_length win_functionfminfmax mel_floorreturn_attention_mask)intfloatboolstr)format __classdict__s"y/home/luca/screen-checking/.venv/lib/python3.14/site-packages/transformers/models/speecht5/feature_extraction_speecht5.py __annotate__%SpeechT5FeatureExtractor.__annotate__Hs+ + + +  +  +  + + + + + + +  $+ c B<\SV`!R RVRVRV/V BW@nWnWPnW`nWpnWnWnWn Wn Wr,R,Vn Wb,R,Vn \VP4VnVP^,^,Vn\!VPVPRR7Vn\%VPVPVPVPVP&RRR7VnR#) rrriT) window_lengthnameperiodicslaney)num_frequency_binsnum_mel_filters min_frequency max_frequencyrnorm mel_scaleN)super__init__rrrrrrrrr sample_size sample_stridern_fftn_freqsrwindowrr mel_filters)selfrrrrrrrrrrrrkwargs __class__s&&&&&&&&&&&&&,r$r5!SpeechT5FeatureExtractor.__init__Hs wlw-w_lwpvw(%:"($$(  "%5='74?'(8(89  a1, %D4D4D4K\K\gkl *#|| --)))),, r'c <V^8dQhRS[S[P,RS[S[P,RS[RS[S[P,/#)rrrrreturn)listnpndarrayr)r"r#s"r$r%r&wsH##2::&#8) rCarrayint32zipsummeansqrtvarshapeappend)rrrnormed_input_valuesvectorlength normed_slicexs&&& r$zero_mean_unit_var_norm0SpeechT5FeatureExtractor.zero_mean_unit_var_normus  %XXnbhh?N"$ "%l4F4Fr4J"K &)=)=)? ?2776RYSY?K^K^K`cgKgChh ..q11,9)#**<8 #L#"Vb"bUaPQLBGGAEEGdN4K#K#KUa "b""#cs AD cN<V^8dQhRS[PRS[P/#)r one_waveformrA)rCrD)r"r#s"r$r%r&s'jj r'c  \VVPVPVPVPVP VP RR7pVP#)zJ Extracts log-mel filterbank features for one waveform array (unbatched). log10)r: frame_lengthr fft_lengthr;rlog_mel)rr:r6r7r8r;rT)r<rX log_mel_specs&& r$_extract_mel_features.SpeechT5FeatureExtractor._extract_mel_featuressP# ;;))))zz((nn  ~~r'c<V^8dQhRS[PS[S[,,S[S[P,,S[S[S[,,,R,RS[PS[S[,,S[S[P,,S[S[S[,,,R,RS[S[,S[,RS[R,RS[RS[R,RS[R,R S[S[,R,R S[R,R S[ / #) raudioN audio_targetpadding max_length truncationpad_to_multiple_ofrreturn_tensorsrrA rCrDrBrr r!r rr r)r"r#s"r$r%r&srrzzDK'$rzz*::T$u+=NNQUUrjj4;.bjj1AADeDUUX\\ro- r $J r  r $Jr $d{rj(4/rTzr rr'c VfVf \R4hV e>WP8wd-\RV RVP RVP RV R2 4hM-\PRVPP R 24VeVP !VR VVVVVV3/V Bp MRp VeFVP !VR VVVVVV3/V Bp V fV #V R ,V R &V PR4p V eWR&V #)a Main method to featurize and prepare for the model one or several sequence(s). Pass in a value for `audio` to extract waveform features. Pass in a value for `audio_target` to extract log-mel spectrogram features. Args: audio (`np.ndarray`, `list[float]`, `list[np.ndarray]`, `list[list[float]]`, *optional*): The sequence or batch of sequences to be processed. Each sequence can be a numpy array, a list of float values, a list of numpy arrays or a list of list of float values. This outputs waveform features. Must be mono channel audio, not stereo, i.e. single float per timestep. audio_target (`np.ndarray`, `list[float]`, `list[np.ndarray]`, `list[list[float]]`, *optional*): The sequence or batch of sequences to be processed as targets. Each sequence can be a numpy array, a list of float values, a list of numpy arrays or a list of list of float values. This outputs log-mel spectrogram features. padding (`bool`, `str` or [`~utils.PaddingStrategy`], *optional*, defaults to `False`): Select a strategy to pad the returned sequences (according to the model's padding side and padding index) among: - `True` or `'longest'`: Pad to the longest sequence in the batch (or no padding if only a single sequence if provided). - `'max_length'`: Pad to a maximum length specified with the argument `max_length` or to the maximum acceptable input length for the model if that argument is not provided. - `False` or `'do_not_pad'` (default): No padding (i.e., can output a batch with sequences of different lengths). max_length (`int`, *optional*): Maximum length of the returned list and optionally padding length (see above). truncation (`bool`): Activates truncation to cut input sequences longer than *max_length* to *max_length*. pad_to_multiple_of (`int`, *optional*): If set will pad the sequence to a multiple of the provided value. This is especially useful to enable the use of Tensor Cores on NVIDIA hardware with compute capability `>= 7.5` (Volta), or on TPUs which benefit from having sequence lengths be a multiple of 128. return_attention_mask (`bool`, *optional*): Whether to return the attention mask. If left to the default, will return the attention mask according to the specific feature_extractor's default. [What are attention masks?](../glossary#attention-mask) return_tensors (`str` or [`~utils.TensorType`], *optional*): If set, will return tensors instead of list of python integers. Acceptable values are: - `'pt'`: Return PyTorch `torch.Tensor` objects. - `'np'`: Return Numpy `np.ndarray` objects. sampling_rate (`int`, *optional*): The sampling rate at which the `audio` or `audio_target` input was sampled. It is strongly recommended to pass `sampling_rate` at the forward call to prevent silent errors. Nz9You must provide either `audio` or `audio_target` values.z3The model corresponding to this feature extractor: z& was trained using a sampling rate of zB. Please make sure that the provided audio input was sampled with z and not .zDIt is strongly recommended to pass the `sampling_rate` argument to `zN()`. Failing to do so can result in silent errors that might be hard to debug.FTrlabelsrdecoder_attention_mask) ValueErrorrloggerwarningr>__name___process_audioget)r<rcrdrerfrgrhrrirr=inputs inputs_targetrns&&&&&&&&&&, r$__call__!SpeechT5FeatureExtractor.__call__sb| =\1XY Y  $ 2 22 I$P**+,**+9]O1F3 NNVW[WeWeWnWnVop\\   (("%  FF  # //"%  M~$$#0#@x )6):):;K)L&)57M34 r'cP<V^8dQhRS[PS[S[,,S[S[P,,S[S[S[,,,RS[RS[S[,S[,RS[R,RS[RS[R,RS[R,R S[S[,R,R S[ / #) rspeech is_targetrerfNrgrhrrirArj)r"r#s"r$r%r&sUU T%[(4 +;;d4;>OOUUo- U $J U  U $JU $d{Uj(4/U Ur'c  \V\P4;'d\VP4^8p V 'd)\VP4^8d\ RV 24hT ;'gK\V\ \34;'d-\V^,\P\\ 34p V 'd8VUu.uF)p\P!V\PR7NK+ uppMV 'gH\V\P4'g(\P!V\PR7pMq\V\P4'dRVP\P!\P4Jd VP\P4pV 'gV.pVPp V'd?VU u.uFqPV 4NK pp \RV/4pVP Vn M \RV/4pVP"!V3RVRVRVRVRV/V BpWn VR,p\V^,\P4'g<VUu.uF)p\P!V\PR7NK+ upVR&EM'\V\P4'g\V^,\P4'dmV^,P\P!\P4Jd4VUu.uF"pVP\P4NK$ upVR&Mt\V\P4'dUVP\P!\P4Jd#VP\P4VR&VP%R 4pVe:VUu.uF)p\P!V\P&R7NK+ upVR &V'gcVP('dQVP+W4R 7\,P.JdTMR pVP1VR,VVP2R 7VR&VeVP5V4pV#uupiuup iuupiuupiuupi) z2Only mono-channel audio is supported for input to )dtyperrerfrgrhrrN)rf)rr) isinstancerCrDlenrNrorBtupleasarrayfloat32r~float64astyperr`rrpadrtrHr_get_padding_strategiesr DO_NOT_PADrUrconvert_to_tensors)r<rzr{rerfrgrhrrir=is_batched_numpy is_batchedfeature_size_hackwaveformfeaturesencoded_inputs padded_inputsrrGrs&&&&&&&&&, r$rs'SpeechT5FeatureExtractor._process_audios&fbjj9SSc&,,>ORS>S FLL 1A 5QRVQWXY Y%  ve} - d d:fQi"**V[]aIb3c  IOPvbjjrzz:PFJvrzz$B$BZZbjj9F  + + @T0T]]2::.FXF!-- MSTV2288*DEN $ 1 1D )>6*BCN   " "  2  #8   .%^4 ,q/2::66^j,k^jUZRZZRZZ-P^j,kM. )<44<?BJJ77Q%%"**)==S_,`S_%U\\"**-ES_,`M. )  bjj 1 1l6H6HBHHUWU_U_L`6`,8,?,? ,KM. )'**+;<  %^l.m^lUZrzz%rxx/P^l.mM* +T...///OWfWqWqq  -1,H,Hn-n\`\n\n-I-M. )  %)<<^LMCQU*-l -a/ns /R,!R1?/R6:(R;6/Sc6<V^8dQhRS[S[S[3,/#)rrA)dictr!r)r"r#s"r$r%r&js  c3h r'cR<\SV`4p.ROpVF pW19gK WK V#)r:)r:r;r6r7r8r9)r4to_dict)r<outputnamesr*r>s& r$r SpeechT5FeatureExtractor.to_dictjs2"^D~L r')rrrrrr;rr8r9rrr6r7rrr:) r}i>FP@ hann_windowrig|=T)r) NNFNFNNNN)FFNFNNN)rr __module__ __qualname____firstlineno____doc__model_input_namesr5 staticmethodrUr`rwrsr__static_attributes____classdictcell__ __classcell__)r>r#s@@r$r r sp&P()9:+ + Z###*&rrhUUn   r'r )rtypingrnumpyrC audio_utilsrrrr!feature_extraction_sequence_utilsrfeature_extraction_utilsrutilsr r r get_loggerrrrpr __all__r3r'r$rsN,\\I499   H %V7Vr & &r'