Add text-text (audio) CLIP #504
Closed
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
|
Oyo, so, where does this sit relative to #323? Is the intent to have that one cleaned up and merged and then this one will merge those changes before being merged? |
|
Hi @rwightman ! I added description, basically we pretend that encoded audio is just text tokens and treat it respectively. I trained it on a small subset of LAION630k (basically VGGSound) for just a little and it gave me not so bad results. I'm just wondering is it even worth it to have such a model and what other people think of it :) |
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Outdated suggestions cannot be applied.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.
Text-Text (audio) CLIP (based on #323 but one of the text tokens are actually audio encoded with Encodec (can add other encoding methods). One of the models is just BERT (for real text) and the other one is LSG BERT.