Repository navigation
fix(scrawl): a capture knows when each word was said - #28
Merged
Merged
Conversation
Canary, the best dictation model, cannot time words, so a capture's whole speech landed in one block at its last mark. A capture's transcript now comes from Parakeet, loaded for that one run on the same backend and let go after. Dictation stays on Canary. If Parakeet is missing or fails, the capture keeps Canary's untimed words as before. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
github-actions
Bot
added
vouch:trusted
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Canary cannot time words (
max_timestamp_kind none), so a capture put all its speech at the last mark. A capture's transcript now comes from Parakeet; dictation stays on Canary. Parakeet is loaded for one run and released, with a fallback to untimed Canary words.Checked with the real engine on a stored 36.5 s capture: 27 timed words on Vulkan0, finish in 0.85 s including the Parakeet load; dictation on the same audio 0.32 s.
Narration still groups by sentence (deliberate), so a recording with little punctuation keeps several marks on one line.
🤖 Generated with Claude Code