mirror of
https://github.com/huggingface/transformers.git
synced 2025-07-04 21:30:07 +06:00
db9dd09cf9
* Adding `AutomaticSpeechRecognitionPipeline`. - Because we added everything to enable this pipeline, we probably should add it to `transformers`. - This PR tries to limit the scope and focuses only on the pipeline part (what should go in, and out). - The tests are very specific for S2T and Wav2vec2 to make sure both architectures are supported by the pipeline. We don't use the mixin for tests right now, because that requires more work in the `pipeline` function (will be done in a follow up PR). - Unsure about the "helper" function `ffmpeg_read`. It makes a lot of sense from a user perspective, it does not add any additional dependencies (as in hard dependency, because users can always use their own load mechanism). Meanwhile, it feels slightly clunky to have so much optional preprocessing. - The pipeline is not done to support streaming audio right now. Future work: - Add `automatic-speech-recognition` as a `task`. And add the FeatureExtractor.from_pretrained within `pipeline` function. - Add small models within tests - Add the Mixin to tests. - Make the logic between ForCTC vs ForConditionalGeneration better. * Update tests/test_pipelines_automatic_speech_recognition.py Co-authored-by: Lysandre Debut <lysandre@huggingface.co> * Adding docs + main import + type checking + LICENSE. * Doc style !. * Fixing TYPE_HINT. * Specifying waveform shape in the docs. * Adding asserts + specify in the documentation the shape of the input np.ndarray. * Update src/transformers/pipelines/automatic_speech_recognition.py Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com> * Adding require to tests + move the `feature_extractor` doc. Co-authored-by: Lysandre Debut <lysandre@huggingface.co> Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>
90 lines
3.6 KiB
Python
90 lines
3.6 KiB
Python
# Copyright 2021 The HuggingFace Team. All rights reserved.
|
|
#
|
|
# Licensed under the Apache License, Version 2.0 (the "License");
|
|
# you may not use this file except in compliance with the License.
|
|
# You may obtain a copy of the License at
|
|
#
|
|
# http://www.apache.org/licenses/LICENSE-2.0
|
|
#
|
|
# Unless required by applicable law or agreed to in writing, software
|
|
# distributed under the License is distributed on an "AS IS" BASIS,
|
|
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
|
# See the License for the specific language governing permissions and
|
|
# limitations under the License.
|
|
|
|
import unittest
|
|
|
|
from transformers import AutoFeatureExtractor, AutoTokenizer, Speech2TextForConditionalGeneration, Wav2Vec2ForCTC
|
|
from transformers.pipelines import AutomaticSpeechRecognitionPipeline
|
|
from transformers.testing_utils import require_datasets, require_torch, require_torchaudio, slow
|
|
|
|
|
|
# from .test_pipelines_common import CustomInputPipelineCommonMixin
|
|
|
|
|
|
class AutomaticSpeechRecognitionPipelineTests(unittest.TestCase):
|
|
# pipeline_task = "automatic-speech-recognition"
|
|
# small_models = ["facebook/s2t-small-mustc-en-fr-st"] # Models tested without the @slow decorator
|
|
# large_models = [
|
|
# "facebook/wav2vec2-base-960h",
|
|
# "facebook/s2t-small-mustc-en-fr-st",
|
|
# ] # Models tested with the @slow decorator
|
|
|
|
@slow
|
|
@require_torch
|
|
@require_datasets
|
|
def test_simple_wav2vec2(self):
|
|
import numpy as np
|
|
from datasets import load_dataset
|
|
|
|
model = Wav2Vec2ForCTC.from_pretrained("facebook/wav2vec2-base-960h")
|
|
tokenizer = AutoTokenizer.from_pretrained("facebook/wav2vec2-base-960h")
|
|
feature_extractor = AutoFeatureExtractor.from_pretrained("facebook/wav2vec2-base-960h")
|
|
|
|
asr = AutomaticSpeechRecognitionPipeline(model=model, tokenizer=tokenizer, feature_extractor=feature_extractor)
|
|
|
|
waveform = np.zeros((34000,))
|
|
output = asr(waveform)
|
|
self.assertEqual(output, {"text": ""})
|
|
|
|
ds = load_dataset("patrickvonplaten/librispeech_asr_dummy", "clean", split="validation")
|
|
filename = ds[0]["file"]
|
|
output = asr(filename)
|
|
self.assertEqual(output, {"text": "A MAN SAID TO THE UNIVERSE SIR I EXIST"})
|
|
|
|
filename = ds[0]["file"]
|
|
with open(filename, "rb") as f:
|
|
data = f.read()
|
|
output = asr(data)
|
|
self.assertEqual(output, {"text": "A MAN SAID TO THE UNIVERSE SIR I EXIST"})
|
|
|
|
@slow
|
|
@require_torch
|
|
@require_torchaudio
|
|
@require_datasets
|
|
def test_simple_s2t(self):
|
|
import numpy as np
|
|
from datasets import load_dataset
|
|
|
|
model = Speech2TextForConditionalGeneration.from_pretrained("facebook/s2t-small-mustc-en-it-st")
|
|
tokenizer = AutoTokenizer.from_pretrained("facebook/s2t-small-mustc-en-it-st")
|
|
feature_extractor = AutoFeatureExtractor.from_pretrained("facebook/s2t-small-mustc-en-it-st")
|
|
|
|
asr = AutomaticSpeechRecognitionPipeline(model=model, tokenizer=tokenizer, feature_extractor=feature_extractor)
|
|
|
|
waveform = np.zeros((34000,))
|
|
|
|
output = asr(waveform)
|
|
self.assertEqual(output, {"text": "E questo è il motivo per cui non ci siamo mai incontrati."})
|
|
|
|
ds = load_dataset("patrickvonplaten/librispeech_asr_dummy", "clean", split="validation")
|
|
filename = ds[0]["file"]
|
|
output = asr(filename)
|
|
self.assertEqual(output, {"text": "Un uomo disse all'universo: \"Signore, io esisto."})
|
|
|
|
filename = ds[0]["file"]
|
|
with open(filename, "rb") as f:
|
|
data = f.read()
|
|
output = asr(data)
|
|
self.assertEqual(output, {"text": "Un uomo disse all'universo: \"Signore, io esisto."})
|