Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Add w2v2bert to pipeline #28585

Merged
merged 3 commits into from
Jan 19, 2024
Merged

Add w2v2bert to pipeline #28585

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
330b3ed
generalize asr pipeline to fbank models
ylacombe Jan 18, 2024
a15e377
change w2v2 pipeline output
ylacombe Jan 18, 2024
d509edc
Update test_pipelines_automatic_speech_recognition.py
ylacombe Jan 19, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
7 changes: 5 additions & 2 deletions src/transformers/pipelines/automatic_speech_recognition.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -582,8 +582,11 @@ def _forward(self, model_inputs, return_timestamps=False, generate_kwargs=None):
out["stride"] = stride

else:
input_values = model_inputs.pop("input_values")
outputs = self.model(input_values=input_values, attention_mask=attention_mask)
inputs = {
self.model.main_input_name: model_inputs.pop(self.model.main_input_name),
"attention_mask": attention_mask,
}
outputs = self.model(**inputs)
logits = outputs.logits

if self.type == "ctc_with_lm":
Expand Down
17 changes: 17 additions & 0 deletions tests/pipelines/test_pipelines_automatic_speech_recognition.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -298,6 +298,23 @@ def test_torch_large(self):
output = speech_recognizer(filename)
self.assertEqual(output, {"text": "A MAN SAID TO THE UNIVERSE SIR I EXIST"})

@require_torch
@slow
def test_torch_large_with_input_features(self):
speech_recognizer = pipeline(
task="automatic-speech-recognition",
model="hf-audio/wav2vec2-bert-CV16-en",
framework="pt",
)
waveform = np.tile(np.arange(1000, dtype=np.float32), 34)
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

it takes mel-spectrograms as input_features, instead of raw waveform as input_values

Is wavefrom the best name here? It looks a bit contradicted to the PR description.

Not a big deal, but if a more proper name is available, we can use it then.

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The input is a waveform but is processed into a melspectrogram (fbank) ! I'll keep this name

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

On that note, I see docstring still says:

input_features (torch.FloatTensor of shape (batch_size, sequence_length)):
Float values of input raw speech waveform.

transformers/src/transformers/models/wav2vec2_bert/modeling_wav2vec2_bert.py

Lines 1020 to 1025 in 708b19e

Args:
input_features (`torch.FloatTensor` of shape `(batch_size, sequence_length)`):
Float values of input raw speech waveform. Values can be obtained by loading a `.flac` or `.wav` audio file
into an array of type `List[float]` or a `numpy.ndarray`, *e.g.* via the soundfile library (`pip install
soundfile`). To prepare the array into `input_features`, the [`AutoProcessor`] should be used for padding and
conversion into a tensor of type `torch.FloatTensor`. See [`Wav2Vec2BertProcessor.__call__`] for details.

output = speech_recognizer(waveform)
self.assertEqual(output, {"text": ""})

ds = load_dataset("hf-internal-testing/librispeech_asr_dummy", "clean", split="validation").sort("id")
filename = ds[40]["file"]
output = speech_recognizer(filename)
self.assertEqual(output, {"text": "a man said to the universe sir i exist"})

@slow
@require_torch
@slow
Expand Down
Loading