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.

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

[MMS] Fix concatenation of emissions #5133

Merged
merged 1 commit into from
May 23, 2023
Merged

Conversation

vineelpratap
Copy link
Contributor

  1. Fix the dimension used to concatenate individual emissions for long audio file
  2. Also fix a bug where context was being removed along label dimension (instead of time dimension)

@vineelpratap vineelpratap merged commit 87d3005 into main May 23, 2023
1 of 7 checks passed
@vineelpratap vineelpratap deleted the vineelpratap-patch-3 branch May 23, 2023 18:35
@yunbin
Copy link

yunbin commented May 23, 2023

#5133 seems fixed the dimension issue, but now have this issue
File "./examples/mms/data_prep/align_and_segment.py", line 90, in get_alignments
path, _ = F.force_align(
AttributeError: module 'torchaudio.functional' has no attribute 'force_align'

@vineelpratap
Copy link
Contributor Author

Please install torchaudio nightly build. See Step 1 here for more details - https://github.com/facebookresearch/fairseq/tree/main/examples/mms/data_prep.

@yunbin
Copy link

yunbin commented May 23, 2023

Yes, the error message was after step 1. conda list show:
torch 2.1.0.dev20230523+cu118 pypi_0 pypi
torchaudio 2.1.0.dev20230523+cu118 pypi_0 pypi

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

Successfully merging this pull request may close these issues.

None yet

3 participants