Hybrid autoregressive transducer (HAT) #1244
Merged
Conversation
This file contains 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
|
Great! |
|
@csukuangfj could you also check this when you have some time? Thanks! |
csukuangfj
reviewed
Dec 6, 2023
Thanks! Left a minor comment. Otherwise, it looks good to me. |
Sorry it took a while since I was on vacation last 2 weeks. I have made the change. |
|
Thanks! |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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.
This is an implementation of the HAT loss proposed in https://arxiv.org/abs/2003.07705.
The test produces reasonable looking losses. I am working on a LibriSpeech zipformer recipe using this loss. In general, it is not expected to improve upon the RNNT loss by itself, but may be useful for things like using external LMs. I am planning to use it in multi-talker ASR for speaker attribution (e.g. https://arxiv.org/abs/2309.08489)