fix kokoro chunker - #172
Merged
Merged
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
In order to keep the inference fast and within the character based context window of kokoro. Internally openarc will chunk the input text meaning you dont need to do anything special to preprocess- just send kokoro your text and the engine takes care of the rest.
Issues were present in how the chunker checks for boundaries- occasionally it would drop text meaning the audio would come through spotty or incomplete. This patch may even improve prosody for this little model.
src/server/schemas/modeling/contract_kokoro.py
• Default character_count_chunk: 100 → 400.
src/server/worker_registry.py
• Removed chunk_texts from Kokoro inference metrics — the full input text was being dumped into server logs on every request.
tests/unit/test_ov_genai_kokoro_unit.py
• 5 new regression tests: overlong-sentence-after-flush splitting, size/lossless invariants, clause-boundary preference, newline/empty-input handling, and multi-bucket audio concatenatio