Aluode/PerceptionLabPortable

1# coding=utf-82# Copyright 2025 The HuggingFace Inc. team. All rights reserved.3#4# Licensed under the Apache License, Version 2.0 (the "License");5# you may not use this file except in compliance with the License.6# You may obtain a copy of the License at7#8#     http://www.apache.org/licenses/LICENSE-2.09#10# Unless required by applicable law or agreed to in writing, software11# distributed under the License is distributed on an "AS IS" BASIS,12# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.13# See the License for the specific language governing permissions and14# limitations under the License.15from typing import Optional, Union16 17from ...audio_utils import AudioInput, make_list_of_audio18from ...processing_utils import ProcessingKwargs, ProcessorMixin, Unpack19from ...tokenization_utils_base import PreTokenizedInput, TextInput20from ...utils import logging21 22 23logger = logging.get_logger(__name__)24 25 26class ParakeetProcessorKwargs(ProcessingKwargs, total=False):27    _defaults = {28        "audio_kwargs": {29            "sampling_rate": 16000,30            "padding": "longest",31        },32        "text_kwargs": {33            "padding": True,34            "padding_side": "right",35            "add_special_tokens": False,36        },37        "common_kwargs": {"return_tensors": "pt"},38    }39 40 41class ParakeetProcessor(ProcessorMixin):42    attributes = ["feature_extractor", "tokenizer"]43    feature_extractor_class = "ParakeetFeatureExtractor"44    tokenizer_class = "ParakeetTokenizerFast"45 46    def __call__(47        self,48        audio: AudioInput,49        text: Union[TextInput, PreTokenizedInput, list[TextInput], list[PreTokenizedInput], None] = None,50        sampling_rate: Optional[int] = None,51        **kwargs: Unpack[ParakeetProcessorKwargs],52    ):53        audio = make_list_of_audio(audio)54 55        output_kwargs = self._merge_kwargs(56            ParakeetProcessorKwargs,57            tokenizer_init_kwargs=self.tokenizer.init_kwargs,58            **kwargs,59        )60 61        if sampling_rate is None:62            logger.warning_once(63                f"You've provided audio without specifying the sampling rate. It will be assumed to be {output_kwargs['audio_kwargs']['sampling_rate']}, which can result in silent errors."64            )65        elif sampling_rate != output_kwargs["audio_kwargs"]["sampling_rate"]:66            raise ValueError(67                f"The sampling rate of the audio ({sampling_rate}) does not match the sampling rate of the processor ({output_kwargs['audio_kwargs']['sampling_rate']}). Please provide resampled the audio to the expected sampling rate."68            )69 70        if audio is not None:71            inputs = self.feature_extractor(audio, **output_kwargs["audio_kwargs"])72        if text is not None:73            encodings = self.tokenizer(text, **output_kwargs["text_kwargs"])74 75        if text is None:76            return inputs77        else:78            inputs["labels"] = encodings["input_ids"]79            return inputs80 81    @property82    def model_input_names(self):83        feature_extractor_input_names = self.feature_extractor.model_input_names84        return feature_extractor_input_names + ["labels"]85 86 87__all__ = ["ParakeetProcessor"]88