[MAEB] Wav2Clip Text Encoder#3781
Closed
AdnanElAssadi56 wants to merge 2 commits intoembeddings-benchmark:maebfrom
Closed
[MAEB] Wav2Clip Text Encoder#3781AdnanElAssadi56 wants to merge 2 commits intoembeddings-benchmark:maebfrom
AdnanElAssadi56 wants to merge 2 commits intoembeddings-benchmark:maebfrom
Conversation
Samoed
reviewed
Dec 22, 2025
| # text side (CLIP) | ||
| self.clip = CLIPModel.from_pretrained(model_name, revision=revision).to(device) | ||
| # text side (CLIP) - we use the standard OpenAI CLIP model as mentioned in paper | ||
| clip_model_name = "openai/clip-vit-base-patch32" |
Member
There was a problem hiding this comment.
Can we process this without loading other model?
Contributor
Author
There was a problem hiding this comment.
If you mean this way:
self.clip = CLIPModel.from_pretrained(model_name, revision=revision).to(device)
Then, no, because there is no huggingface revision for the model, and it was giving a 404 error during evaluation.
Member
There was a problem hiding this comment.
I mean to retrieve from existing model somehow, but they're using this method for encoding and this is hard to tell how to do this with this model https://github.com/descriptinc/lyrebird-wav2clip/blob/1864b3924be5a785e2d49d975b8a26ff93f62951/wav2clip/__init__.py#L34
Samoed
reviewed
Dec 22, 2025
Contributor
|
This pull request has been automatically marked as stale due to inactivity. |
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Related to #3545
@Samoed I've double-checked the paper (arXiv:2110.11499v2), and it seem to confirm that we can use the standard CLIP text encoder.
Since the text encoder is identical to the standard CLIP encoder, I think we can safely get the text embeddings from the original CLIP model, and they will be mathematically aligned with the audio embeddings from wav2clip.