Skip to content

Conversation

@anton-l
Copy link

@anton-l anton-l commented Oct 13, 2021

This PR updates the padding mask calculation to be the same as the one in the reference Wav2Vec2 implementation (same commit as listed in SEW's README): https://github.com/pytorch/fairseq/blob/05255f96410e5b1eaf3bf59b767d5b4b7e2c3a35/fairseq/models/wav2vec/wav2vec2.py#L477

For more details on how and why it was fixed in fairseq, check out this PR by @patrickvonplaten facebookresearch/fairseq#3228

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant