Do not encode supplied initial_prompt if it is not a string#1299
Open
jzyeezy wants to merge 4 commits into
Open
Do not encode supplied initial_prompt if it is not a string#1299jzyeezy wants to merge 4 commits into
initial_prompt if it is not a string#1299jzyeezy wants to merge 4 commits into
Conversation
MahmoudAshraf97
requested changes
May 19, 2025
Collaborator
MahmoudAshraf97
left a comment
There was a problem hiding this comment.
Thanks for your contribution, You still need to handle the case where options.initial_prompt is None
Collaborator
|
Also, even if the typing hints support, Why would you use a tokenized string instead of a string? note that different whisper models use different tokenizers so the tokenized string is not generalizable |
Author
@MahmoudAshraf97 gotcha. I only discovered this bug while experimenting with the parameter to observe if there's any upside in providing the prompt in this format. I'm happy to update the signature of this parameter to accept only a |
This reverts commit 3c425b6.
…lues for `initial_prompt`
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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
BatchedInferencePipeline#transcribemethod technically supports passing an already encoded tokenized list as theinitial_prompt. The code path should only try to encode the provided prompt if it's a string.This fixes the bug where faster_whisper will fail due to:
TypeError: TextInputSequence must be strwhen passing a List of integers as theinitial_prompt.