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

Basque alignment model #1074

Merged
merged 2 commits into from
Mar 4, 2025
Merged

Basque alignment model #1074

merged 2 commits into from
Mar 4, 2025

Conversation

xezpeleta
Copy link
Contributor

The following PR adds a phoneme-based alignment model for Basque language.

As a result we obtain high-precision word-level timestamps.

Demo: https://youtu.be/adB62l_PdYA

Thanks

Copy link
Collaborator

@Barabazs Barabazs left a comment

Choose a reason for hiding this comment

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

Thank you!

@Barabazs Barabazs merged commit 0d9807a into m-bain:main Mar 4, 2025
@jim60105
Copy link
Contributor

jim60105 commented Mar 4, 2025

@Barabazs We now have two 'eu.' Is this actually working?

@xezpeleta
Copy link
Contributor Author

Sorry, I didn't realize there was already an "eu" in the list. I'm going to add a new PR to fix it.

Barabazs added a commit to Barabazs/whisperX that referenced this pull request Mar 5, 2025
Barabazs added a commit that referenced this pull request Mar 5, 2025
@Barabazs
Copy link
Collaborator

Barabazs commented Mar 5, 2025

@Barabazs We now have two 'eu.' Is this actually working?

Thanks for catching this. Reverted in #1077

@xezpeleta
Copy link
Contributor Author

Ok, I've just tested it with two different alignment models, and I don't see any difference in alignment accuracy. So, I'm not going to suggest any changes to the current configuration.

Thanks, and sorry for the confusion!

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.

3 participants