o wi@sdZddlZddlmZddlmZmZddlmZm Z m Z ddl m Z m Z mZdd lmZdd lmZGd d d ed dZGddde ZdgZdS)z% Speech processor class for Wav2Vec2 N)contextmanager)OptionalUnion)ProcessingKwargsProcessorMixinUnpack) AudioInputPreTokenizedInput TextInput)Wav2Vec2FeatureExtractor)Wav2Vec2CTCTokenizerc@seZdZiZdS)Wav2Vec2ProcessorKwargsN)__name__ __module__ __qualname__ _defaultsrrm/home/ubuntu/sommelier/.venv/lib/python3.10/site-packages/transformers/models/wav2vec2/processing_wav2vec2.pyrsrF)totalc seZdZdZdZdZfddZefddZ    dd e d e e e e e eefd eefd d ZddZddZddZeddZZS)Wav2Vec2Processora Constructs a Wav2Vec2 processor which wraps a Wav2Vec2 feature extractor and a Wav2Vec2 CTC tokenizer into a single processor. [`Wav2Vec2Processor`] offers all the functionalities of [`Wav2Vec2FeatureExtractor`] and [`PreTrainedTokenizer`]. See the docstring of [`~Wav2Vec2Processor.__call__`] and [`~Wav2Vec2Processor.decode`] for more information. Args: feature_extractor (`Wav2Vec2FeatureExtractor`): An instance of [`Wav2Vec2FeatureExtractor`]. The feature extractor is a required input. tokenizer ([`PreTrainedTokenizer`]): An instance of [`PreTrainedTokenizer`]. The tokenizer is a required input. r AutoTokenizercs t|||j|_d|_dS)NF)super__init__feature_extractorcurrent_processor_in_target_context_manager)selfr tokenizer __class__rrr3s zWav2Vec2Processor.__init__c stz tj|fi|WSttfy9td|jdttj|fi|}t j|fi|}|||dYSw)NzLoading a tokenizer inside a from a config that does not include a `tokenizer_class` attribute is deprecated and will be removed in v5. Please add `'tokenizer_class': 'Wav2Vec2CTCTokenizer'` attribute to either your `config.json` or `tokenizer_config.json` file to suppress this warning: )rr) rfrom_pretrainedOSError ValueErrorwarningswarnr FutureWarningr r)clspretrained_model_name_or_pathkwargsrrr rrr"8s  z!Wav2Vec2Processor.from_pretrainedNaudiotextr*c Ksd|vrtd|d}|dur|durtd|jtfd|jji|}|jr=|j |fi|d|d|dS|durL|j |fi|d}|dur[|j|fi|d}|dura|S|durg|S|d |d <|S) z This method forwards all its arguments to Wav2Vec2FeatureExtractor's [`~Wav2Vec2FeatureExtractor.__call__`] and returns its output. raw_speechzLUsing `raw_speech` as a keyword argument is deprecated. Use `audio` instead.NzAYou need to specify either an `audio` or `text` input to process.tokenizer_init_kwargs audio_kwargs text_kwargs common_kwargs input_idslabels) r%r&popr$ _merge_kwargsrr init_kwargsrrr) rr+r,imagesvideosr* output_kwargsinputs encodingsrrr__call__Ks@   zWav2Vec2Processor.__call__cOs|jr |jj|i|S|dd}|dd}t|dkr(|d}|dd}|dur9|jj|g|Ri|}|durG|jj|fi|}|durM|S|durS|S|d|d<|S)z This method forwards all its arguments to Wav2Vec2FeatureExtractor's [`~Wav2Vec2FeatureExtractor.pad`] and returns its output. input_featuresNr3rr r2)rrpadr4lenrr)rargsr*r=r3rrrr>zs"     zWav2Vec2Processor.padcO|jj|i|S)z This method forwards all its arguments to PreTrainedTokenizer's [`~PreTrainedTokenizer.batch_decode`]. Please refer to the docstring of this method for more information. )r batch_decoderr@r*rrrrBzWav2Vec2Processor.batch_decodecOrA)z This method forwards all its arguments to PreTrainedTokenizer's [`~PreTrainedTokenizer.decode`]. Please refer to the docstring of this method for more information. )rdecoderCrrrrErDzWav2Vec2Processor.decodeccs2tdd|_|j|_dV|j|_d|_dS)z Temporarily sets the tokenizer for processing the input. Useful for encoding the labels when fine-tuning Wav2Vec2. z`as_target_processor` is deprecated and will be removed in v5 of Transformers. You can process your labels by using the argument `text` of the regular `__call__` method (either in the same call as your audio inputs, or in a separate call.TNF)r%r&rrrr)rrrras_target_processors z%Wav2Vec2Processor.as_target_processor)NNNN)rrr__doc__feature_extractor_classtokenizer_classr classmethodr"r rrstrlistr r rrr<r>rBrErrF __classcell__rrr rr!s.  /r)rGr% contextlibrtypingrrprocessing_utilsrrrtokenization_utils_baser r r feature_extraction_wav2vec2r tokenization_wav2vec2rrr__all__rrrrs