fix: respect language parameter in vLLM transcription API#101
Open
gordon0414 wants to merge 1 commit intoQwenLM:mainfrom
Open
fix: respect language parameter in vLLM transcription API#101gordon0414 wants to merge 1 commit intoQwenLM:mainfrom
gordon0414 wants to merge 1 commit intoQwenLM:mainfrom
Conversation
When using the vLLM /v1/audio/transcriptions endpoint, the `language` parameter was ignored because `get_generation_prompt` only checked `to_language` and never fell back to `language`. This adds a fallback so that when `to_language` is None, the `language` parameter is used instead, allowing the Whisper-compatible `language` field to correctly inject the language prefill into the assistant prompt. Fixes: vllm-project/vllm#33768
1 task
|
@wangxiongts Please take some time to review and handle this merge request. |
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
When using the vLLM /v1/audio/transcriptions endpoint, the
languageparameter was ignored becauseget_generation_promptonly checkedto_languageand never fell back tolanguage.This adds a fallback so that when
to_languageis None, thelanguageparameter is used instead, allowing the Whisper-compatiblelanguagefield to correctly inject the language prefill into the assistant prompt.Fixes: vllm-project/vllm#33768
Testing
Tested with Qwen3-ASR-1.7B on vLLM serving (4 languages):
language=ko, JA audiolanguage=ko, KO audio