Add support for fixed character-level StreamAtt history selection#40
Open
sarapapi wants to merge 4 commits into
Open
Add support for fixed character-level StreamAtt history selection#40sarapapi wants to merge 4 commits into
sarapapi wants to merge 4 commits into
Conversation
Contributor
|
LGTM apart one minor comment, thanks |
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
FixedWordsTextHistorycounts words using the BOW prefix (▁) to identify word boundaries, which works well for space-separated languages but not for character-level languages (e.g., Chinese, Japanese) where BOW markers are sparse. When few tokens carry a BOW prefix, the word counter never reacheshistory_words, the text history is never trimmed, and the audio buffer grows unboundedly, causing AlignAtt to progressively cut all newly generated tokens.This PR extends the FixedWords-style approach to character-level languages by introducing
FixedCharsTextHistory, which counts every token as one unit (mirroring how character-level segmenters treat individual CJK characters). A dedicated config file (seamless_streamatt_char.yaml) is provided, pairingFixedCharsTextHistorywithword_level_postprocess: False, required to prevent_strip_incomplete_wordsfrom discarding valid CJK tokens.