o ZŽhJã@s>dZddlZddlmZddlmZGdd„deƒZdgZdS)z$ Speech processor class for M-CTC-T éN)Úcontextmanageré)ÚProcessorMixincsTeZdZdZdZdZ‡fdd„Zdd„Zdd „Zd d „Z d d „Z e dd„ƒZ ‡Z S)ÚMCTCTProcessora[ Constructs a MCTCT processor which wraps a MCTCT feature extractor and a MCTCT tokenizer into a single processor. [`MCTCTProcessor`] offers all the functionalities of [`MCTCTFeatureExtractor`] and [`AutoTokenizer`]. See the [`~MCTCTProcessor.__call__`] and [`~MCTCTProcessor.decode`] for more information. Args: feature_extractor (`MCTCTFeatureExtractor`): An instance of [`MCTCTFeatureExtractor`]. The feature extractor is a required input. tokenizer (`AutoTokenizer`): An instance of [`AutoTokenizer`]. The tokenizer is a required input. ZMCTCTFeatureExtractorZ AutoTokenizercs tƒ ||¡|j|_d|_dS)NF)ÚsuperÚ__init__Úfeature_extractorÚcurrent_processorÚ_in_target_context_manager)ÚselfrÚ tokenizer©Ú __class__©úd/var/www/auris/lib/python3.10/site-packages/transformers/models/deprecated/mctct/processing_mctct.pyr*s zMCTCTProcessor.__init__cOsö|jr |j|i|¤ŽSd|vrt d¡| d¡}n| dd¡}| dd¡}| dd¡}t|ƒdkr<|d}|dd…}|durH|durHtd ƒ‚|durZ|j|g|¢Rd|i|¤Ž}|durg|j|fi|¤Ž}|durm|S|durs|S|d |d <|S) a¤ When used in normal mode, this method forwards all its arguments to MCTCTFeatureExtractor's [`~MCTCTFeatureExtractor.__call__`] and returns its output. If used in the context [`~MCTCTProcessor.as_target_processor`] this method forwards all its arguments to AutoTokenizer's [`~AutoTokenizer.__call__`]. Please refer to the docstring of the above two methods for more information. Z raw_speechzLUsing `raw_speech` as a keyword argument is deprecated. Use `audio` instead.ÚaudioNÚ sampling_rateÚtextrézAYou need to specify either an `audio` or `text` input to process.Ú input_idsÚlabels) r r ÚwarningsÚwarnÚpopÚlenÚ ValueErrorrr )r ÚargsÚkwargsrrrZinputsÚ encodingsrrrÚ__call__/s.        zMCTCTProcessor.__call__cOó|jj|i|¤ŽS)z½ This method forwards all its arguments to AutoTokenizer's [`~PreTrainedTokenizer.batch_decode`]. Please refer to the docstring of this method for more information. )r Ú batch_decode©r rrrrrr!UózMCTCTProcessor.batch_decodecOs¶|jr |jj|i|¤ŽS| dd¡}| dd¡}t|ƒdkr(|d}|dd…}|dur9|jj|g|¢Ri|¤Ž}|durG|jj|fi|¤Ž}|durM|S|durS|S|d|d<|S)a¦ When used in normal mode, this method forwards all its arguments to MCTCTFeatureExtractor's [`~MCTCTFeatureExtractor.pad`] and returns its output. If used in the context [`~MCTCTProcessor.as_target_processor`] this method forwards all its arguments to PreTrainedTokenizer's [`~PreTrainedTokenizer.pad`]. Please refer to the docstring of the above two methods for more information. Úinput_featuresNrrrr)r r Úpadrrrr )r rrr$rrrrr%\s"     zMCTCTProcessor.padcOr )z· This method forwards all its arguments to AutoTokenizer's [`~PreTrainedTokenizer.decode`]. Please refer to the docstring of this method for more information. )r Údecoder"rrrr&zr#zMCTCTProcessor.decodeccs2t d¡d|_|j|_dV|j|_d|_dS)z Temporarily sets the tokenizer for processing the input. Useful for encoding the labels when fine-tuning MCTCT. zî`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)rrr r r r)r rrrÚas_target_processors€ÿ z"MCTCTProcessor.as_target_processor)Ú__name__Ú __module__Ú __qualname__Ú__doc__Zfeature_extractor_classZtokenizer_classrrr!r%r&rr'Ú __classcell__rrr rrs  &r)r+rÚ contextlibrZprocessing_utilsrrÚ__all__rrrrÚs    y