whisper: add carry_initial_prompt to maintain context over sliding wi…#1414
Open
Sahith59 wants to merge 2 commits intoml-explore:mainfrom
Open
whisper: add carry_initial_prompt to maintain context over sliding wi…#1414Sahith59 wants to merge 2 commits intoml-explore:mainfrom
Sahith59 wants to merge 2 commits intoml-explore:mainfrom
Conversation
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Description
This PR resolves the issue where context provided by the user's
initial_promptis lost/truncated as the cross-attention context window slides forward over long audio files.Changes
carry_initial_promptkeyword boolean to.transcribe().transcribe.py. When enabled, it dynamically prepends theinitial_prompt_tokensdirectly to the active prompt queue.model.dims.n_text_ctxto correctly slice theprevious_tokensso that the model evaluates without hitting dimension mismatch errors during Apple MLX evaluation.test.pyto test context-window bounds and propagation.carry_initial_prompt=False).Testing
Ran local conversion via
setUpClassand verified native propagation logic inside the test suite. Formatted withblackand passespre-commitnatively.Resolves: #1410