Fix truncated words list when the replacement character is decoded #1089
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
|
Thanks! Great catches |
zackees
pushed a commit
to zackees/whisper
that referenced
this pull request
May 5, 2023
ilanit1997
pushed a commit
to ilanit1997/whisper
that referenced
this pull request
May 16, 2023
abyesilyurt
pushed a commit
to abyesilyurt/whisper
that referenced
this pull request
Nov 13, 2023
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Outdated suggestions cannot be applied.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.
The current version of method
split_tokens_on_unicodekeeps aggregating tokens while the decoded sequence contains the replacement character �.However, the model can sometimes produce invalid sequences which would be decoded to the replacement character (for example in the temperature fallback). In this case we should include the replacement character in the list of words.
In this PR I propose to save the replacement character if it appears at the same position in the fully decoded sequence.