fix: apply channel averaging correctly in audio feature extractors#45549
Open
jonghwanhyeon wants to merge 1 commit into
Open
fix: apply channel averaging correctly in audio feature extractors#45549jonghwanhyeon wants to merge 1 commit into
jonghwanhyeon wants to merge 1 commit into
Conversation
Contributor
|
[For maintainers] Suggested jobs to run (before merge) run-slow: cohere_asr, lasr, parakeet, phi4_multimodal, voxtral_realtime |
eustlb
reviewed
Apr 21, 2026
eustlb
left a comment
Collaborator
There was a problem hiding this comment.
Hey thanks a lot for your PR, there is indeed a bug here, but this bug is rather due to ambiguous representation of stereo audio across the lib.
Your fix is correct in the following assumption: stereo audio is in shape [BS, channels, audio lenght]. Were we to pass a [channels, audio lenght], then there is confusion between batch size and channel dim.
Anyway, this is unclear and should be covered by #44394, so I am going to alt review here just a bit to think about the best way to solve these ambiguities and come back to you asap.
27 tasks
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
What does this PR do?
This PR fixes incorrect channel averaging in the following audio feature extractors:
CohereAsrFeatureExtractorParakeetFeatureExtractorPhi4MultimodalFeatureExtractorLasrFeatureExtractorVoxtralRealtimeFeatureExtractorThe bug had two parts:
(batch, channels, length)were averaging the last dimension instead of the channel dimension.speech.mean(...)without writing the result back intoraw_speech.This change averages batched tensor inputs across channels with
mean(1)and writes the mono-converted sample back toraw_speechfor sequence inputs.Code Agent Policy
The Transformers repo is currently being overwhelmed by a large number of PRs and issue comments written by
code agents. We are currently bottlenecked by our ability to review and respond to them. As a result,
we ask that new users do not submit pure code agent PRs at this time.
You may use code agents in drafting or to help you diagnose issues. We'd also ask autonomous "OpenClaw"-like agents
not to open any PRs or issues for the moment.
PRs that appear to be fully agent-written will probably be closed without review, and we may block users who do this
repeatedly or maliciously.
This is a rapidly-evolving situation that's causing significant shockwaves in the open-source community. As a result,
this policy is likely to be updated regularly in the near future. For more information, please read
CONTRIBUTING.md.Before submitting
Pull Request section?
to it if that's the case.
documentation guidelines, and
here are tips on formatting docstrings.
Who can review?
@eustlb @ebezzam @vasqu