sg+>dZddlZddlmZddlmZGddeZy)z$ Speech processor class for M-CTC-T N)contextmanager)ProcessorMixincReZdZdZdZdZfdZdZdZdZ dZ e d Z xZ S) MCTCTProcessora[ 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. MCTCTFeatureExtractor AutoTokenizercVt||||j|_d|_y)NF)super__init__feature_extractorcurrent_processor_in_target_context_manager)selfr tokenizer __class__s h/var/www/html/venv/lib/python3.12/site-packages/transformers/models/deprecated/mctct/processing_mctct.pyr zMCTCTProcessor.__init__*s) *I6!%!7!7*/'c|jr|j|i|Sd|vr'tjd|j d}n|j dd}|j dd}|j dd}t |dkDr |d}|dd}| | t d ||j|g|d|i|}||j|fi|}|S|Sd 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 doctsring of the above two methods for more information. raw_speechzLUsing `raw_speech` as a keyword argument is deprecated. Use `audio` instead.audioN sampling_ratetextrzAYou need to specify either an `audio` or `text` input to process. input_idslabels) rrwarningswarnpoplen ValueErrorr r)rargskwargsrrrinputs encodingss r__call__zMCTCTProcessor.__call__/s  * *)4))4:6: : 6 ! MMh iJJ|,EJJw-E ?D9 zz&$' t9q=GE8D =T\`a a  +T++E`D` `Y_`F  &t6v6I <M ] (5F8 Mrc:|jj|i|S)z This method forwards all its arguments to AutoTokenizer's [`~PreTrainedTokenizer.batch_decode`]. Please refer to the docstring of this method for more information. )r batch_decoderr"r#s rr(zMCTCTProcessor.batch_decodeUs +t~~**D;F;;rcp|jr|jj|i|S|jdd}|jdd}t |dkDr |d}|dd}| |j j|g|i|}||j j|fi|}||S||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. input_featuresNrrrr)rrpadrr r r)rr"r#r+rs rr,zMCTCTProcessor.pad\s  * *-4))--t>v> >$4d;Hd+ t9q=!!WN8D  %7T3377XXQWXN  'T^^''9&9F >! !  #M'-k':N8 $! !rc:|jj|i|S)z This method forwards all its arguments to AutoTokenizer's [`~PreTrainedTokenizer.decode`]. Please refer to the docstring of this method for more information. )rdecoder)s rr.zMCTCTProcessor.decodezs %t~~$$d5f55rc#Ktjdd|_|j|_d|j |_d|_yw)z Temporarily sets the tokenizer for processing the input. Useful for encoding the labels when fine-tuning MCTCT. 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)rrrrrr )rs ras_target_processorz"MCTCTProcessor.as_target_processorsH   8 +/'!% !%!7!7*/'sA A )__name__ __module__ __qualname____doc__feature_extractor_classtokenizer_classr r&r(r,r.rr0 __classcell__)rs@rrrsC 6%O0 $L<"<6 0 0rr)r4r contextlibrprocessing_utilsrrrrr;s#%/v0^v0r