Fix Whisper shortform EOS #33917
Closed
Fix Whisper shortform EOS #33917
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
ylacombe
commented
Oct 3, 2024
Comment on lines
+894
to
+900
| if is_shortform and generation_config.pad_token_id == generation_config.eos_token_id: | ||
| num_paddings -= 1 | ||
|
|
||
| if num_paddings != 0: | ||
| seek_sequence = seek_sequence[:-num_paddings] | ||
| if return_token_timestamps and not is_shortform: | ||
| seek_outputs[i]["token_timestamps"] = seek_outputs[i]["token_timestamps"][:-num_paddings] |
There was a problem hiding this comment.
I'm actually not sure if this fix covers all use-cases, @eustlb, would you mind do a sanity check ?
There was a problem hiding this comment.
Let's not merge before being sure
|
The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update. |
4 tasks
|
Closing as it was fixed in #34135 ! 🤗 |
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Outdated suggestions cannot be applied.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.
What does this PR do?
Since short and longform merging, Whisper removed EOS tokens when doing shortform transcription, which is something not happening in the original implementation. It fixes the
test_default_multilingual_transcription_short_formandtest_generate_with_prompt_idstestsA side effect is that average logprob was miscomputed.
cc @eustlb