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

fix: compute fbank on selected device #1529

Merged
merged 2 commits into from
Nov 7, 2023
Merged

fix: compute fbank on selected device #1529

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
97ea695
fix: compute fbank on selected device
hbredin Nov 5, 2023
29a8f51
doc: update changelog
hbredin Nov 7, 2023
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
1 change: 1 addition & 0 deletions CHANGELOG.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -8,6 +8,7 @@
- BREAKING(pipeline): remove `logging_hook` (use `ArtifactHook` instead)
- fix(pipeline): add missing "embedding" hook call in `SpeakerDiarization`
- feat(utils): add `"soft"` option to `Powerset.to_multilabel`
- improve(pipeline): compute `fbank` on GPU when requested

## Version 3.0.1 (2023-09-28)

Expand Down
7 changes: 4 additions & 3 deletions pyannote/audio/pipelines/speaker_verification.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -556,6 +556,7 @@ def compute_fbank(
for waveform in waveforms
]
)

return features - torch.mean(features, dim=1, keepdim=True)

def __call__(
Expand All @@ -578,12 +579,12 @@ def __call__(
batch_size, num_channels, num_samples = waveforms.shape
assert num_channels == 1

features = self.compute_fbank(waveforms)
features = self.compute_fbank(waveforms.to(self.device))
_, num_frames, _ = features.shape

if masks is None:
embeddings = self.session_.run(
output_names=["embs"], input_feed={"feats": features.numpy()}
output_names=["embs"], input_feed={"feats": features.numpy(force=True)}
)[0]

return embeddings
Expand All @@ -606,7 +607,7 @@ def __call__(

embeddings[f] = self.session_.run(
output_names=["embs"],
input_feed={"feats": masked_feature.numpy()[None]},
input_feed={"feats": masked_feature.numpy(force=True)[None]},
)[0][0]

return embeddings
Expand Down
Loading