+
j, ~ R t ^ RIt^RIHt ^RIHt ^RIHtH t H
t
]
P ! ] 4 t
! R R]4 tR.tR# )z&
Feature extractor class for Wav2Vec2
N)SequenceFeatureExtractor) BatchFeature)PaddingStrategy
TensorTypeloggingc x a a ] tR t^ t oRtRR.tR
V 3R llt]R V3R lR ll4 tR V3R lR llt R t
Vt V ;t # )
Wav2Vec2FeatureExtractora
Constructs a Wav2Vec2 feature extractor.
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 `True`):
Whether or not to zero-mean unit-variance normalize the input. Normalizing can help to significantly
improve the performance for some models, *e.g.*,
[wav2vec2-lv60](https://huggingface.co/models?search=lv60).
return_attention_mask (`bool`, *optional*, defaults to `False`):
Whether or not [`~Wav2Vec2FeatureExtractor.__call__`] should return `attention_mask`.
Wav2Vec2 models that have set `config.feat_extract_norm == "group"`, such as
[wav2vec2-base](https://huggingface.co/facebook/wav2vec2-base-960h), have **not** been trained using
`attention_mask`. For such models, `input_values` should simply be padded with 0 and no `attention_mask`
should be passed.
For Wav2Vec2 models that have set `config.feat_extract_norm == "layer"`, such as
[wav2vec2-lv60](https://huggingface.co/facebook/wav2vec2-large-960h-lv60-self), `attention_mask` should be
passed for batched inference.
input_valuesattention_maskc L < \ SV ` ! RR VRVRV/VB W@n WPn R# ) feature_size
sampling_rate
padding_valueN )super__init__return_attention_mask do_normalize)selfr r
r r r kwargs __class__s &&&&&&,y/home/luca/screen-checking/.venv/lib/python3.14/site-packages/transformers/models/wav2vec2/feature_extraction_wav2vec2.pyr !Wav2Vec2FeatureExtractor.__init__@ s2 wlw-w_lwpvw%:" ( c < V ^8 d QhRS[ S[P , RS[ S[P , RS[RS[ S[P , /# ) r r
r return)listnpndarrayfloat)format
__classdict__s "r __annotate__%Wav2Vec2FeatureExtractor.__annotate__N sH # #2::)
r arrayint32zipsummeansqrtvarshapeappend)r r
r normed_input_valuesvectorlength normed_slicexs &&& r zero_mean_unit_var_norm0Wav2Vec2FeatureExtractor.zero_mean_unit_var_normM s % XXnbhh ?N"$ "%l4F4Fr4J"K & )=)=)? ?2776RYSY?K^K^K`cgKgCh h . .q 11,9)#**<8 #L #" Vb"bUaPQLBGGAEEGdN4K#K#KUa "b"" #cs
AD c ^ < V ^8 d QhRS[ P S[S[, , S[S[ P , , S[S[S[, , , RS[S[, S[, RS[R, RS[RS[R, RS[R, RS[S[, R, R S[R, R
S[ / # ) r
raw_speechpadding
max_lengthN
truncationpad_to_multiple_ofr return_tensorsr
r )
r r r r boolstrr intr r )r! r" s "r r# r$ c s I I JJe,tBJJ/??$tE{BSSI o-I $J I
I $J
I $d{I j(4/I TzI
I r c Ve> WP 8w d- \ RV RV P RV P RV R2 4 hM-\ P RV P P
R24 \
V\ P 4 ;' d \ VP 4 ^8 p
V
' d) \ VP 4 ^8 d \ R V 24 hT
;' gK \
V\ \ 34 ;' d- \
V^ , \ P \ \ 34 p V ' g V.p\ R
V/4 p V P V VVVVVR 7 p
V
R
, p\
V^ , \ P 4 ' g< V Uu. uF) p\ P ! V\ P R 7 NK+ upV
R
& EM&\
V\ P 4 ' g \
V^ , \ P 4 ' dl V^ , P" \ P" ! \ P$ 4 J d3 V Uu. uF! qP' \ P 4 NK# upV
R
& Mt\
V\ P 4 ' dU VP" \ P" ! \ P$ 4 J d# VP' \ P 4 V
R
& V
P) R
4 pVe: V Uu. uF) p\ P ! V\ P* R 7 NK+ upV
R
& V P, ' dQ V P/ W#R7 \0 P2 Jd TMRpV P5 V
R
, VV P6 R7 V
R
& Ve V
P9 V4 p
V
# u upi u upi u upi )a5
Main method to featurize and prepare for the model one or several sequence(s).
Args:
raw_speech (`np.ndarray`, `list[float]`, `list[np.ndarray]`, `list[list[float]]`):
The sequence or batch of sequences to be padded. 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. Must be mono channel audio, not
stereo, i.e. single float per timestep.
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)
Wav2Vec2 models that have set `config.feat_extract_norm == "group"`, such as
[wav2vec2-base](https://huggingface.co/facebook/wav2vec2-base-960h), have **not** been trained using
`attention_mask`. For such models, `input_values` should simply be padded with 0 and no
`attention_mask` should be passed.
For Wav2Vec2 models that have set `config.feat_extract_norm == "layer"`, such as
[wav2vec2-lv60](https://huggingface.co/facebook/wav2vec2-large-960h-lv60-self), `attention_mask` should
be passed for batched inference.
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 `raw_speech` input was sampled. It is strongly recommended to pass
`sampling_rate` at the forward call to prevent silent errors.
padding_value (`float`, *optional*, defaults to 0.0):
Nz3The model corresponding to this feature extractor: z& was trained using a sampling rate of zI. Please make sure that the provided `raw_speech` 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.z2Only mono-channel audio is supported for input to r )r9 r: r; r< r )dtyper
)r: )r
r ) r
ValueErrorloggerwarningr __name__
isinstancer r lenr. r tupler padasarrayfloat32rC float64astypegetr( r _get_padding_strategiesr
DO_NOT_PADr5 r convert_to_tensors)r r8 r9 r: r; r< r r= r
r is_batched_numpy
is_batchedencoded_inputs
padded_inputsr r' r
s &&&&&&&&&, r __call__!Wav2Vec2FeatureExtractor.__call__c sI J $ 2 22 I$ P **+ , **+9]O1F 3
NNVW[WeWeWnWnVo p\ \
&j"**=[[#jFVFVBWZ[B[ J$4$4 5 9 QRVQW XY Y%
zD%= 1 l lz*Q-RTR\R\^ceiQj7k
$ J &~z&BC !! 1"7
!
%^4 ,q/2::66^j,k^jUZRZZRZZ-P^j,kM. )<44<?BJJ77 Q%%"**)==S_,`S_%\\"**-ES_,`M. )
bjj
1
1l6H6HBHHUWU_U_L`6`,8,?,?
,KM. ) '**+;< %^l.m^lUZrzz%rxx/P^l.mM* + ///OWfWqWqq
-1,H,H n-n\`\n\n -I -M. ) % ) < <^ LM ; -l -a /ns /N4 'N9/N>)r r ) i> FT)r[ )FNFNNNN)
rG
__module__ __qualname____firstlineno____doc__model_input_namesr staticmethodr5 rX __static_attributes____classdictcell__
__classcell__)r r" s @@r r r sI B ()9: ) # # #*I I I r r )r_ numpyr !feature_extraction_sequence_utilsr feature_extraction_utilsr utilsr r r
get_loggerrG rE r __all__r r r rk sG I 4 9 9
H %P 7 P f &
&r