Repository navigation
Conversation
A caption was a fixed group of words, shown from its first word's start to its last word's end, so a pause inside the group kept it on screen. - A pause of more than 0.8 s ends the caption; the next word starts a new one. - A word counts for at most 1.2 s after its start, so a word the transcript stretched over a silence no longer holds a caption up. - A word timed with no length beside a pause gets a 0.4 s caption of its own instead of holding the caption through the pause. - The editor's copy of the grouping (speakerTurns.ts) does the same; the Python/TypeScript parity tests cover pauses and long words. - Lines saved by hand keep their times. No setting and no new UI text. - With no pause over 0.8 s and no word over 1.2 s the output is byte-identical: no snapshot changed and the golden renders match. Checked: QC and a burned-frame review, one fix round, both re-run afterwards and passed. Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
|
The latest updates on your projects. Learn more about Vercel for GitHub.
1 Skipped Deployment
|
The name was built from the process and thread ids in hex, which on Linux is longer than the font's own file name, so the step 3 test for long clip folders failed in CI. It is now 13 characters everywhere. Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
|
Hey Collin ich hätte ein paar Fragen? Wie erreicht man dich am besten |
Owner
Author
|
I check github issues everyday and I have my social media links at the bottom of the github |
This branch was successfully deployed
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Closes #228. Stacked on #230 (editor suite step 3).
The problem
A caption was a fixed group of words (3 by default), shown from its first word's start to its last word's end. When someone paused inside a group, the caption stayed on screen for the whole pause. A word the transcript stretched over a silence did the same.
The fix
It is the default behaviour, with no setting and no new UI text.
Measured
On a real 150-second transcript, burned and measured frame by frame:
Caption time on screen while nobody is speaking fell from about 31 s to about 7 s over 113 s of video (3 words per caption). What remains is breaths shorter than 0.8 s.
What stays the same
Where there is no pause over 0.8 s and no word over 1.2 s, the output is byte-identical: 17,898 line and ASS comparisons, 132 burned clips with identical frames, no snapshot changed, and the four golden renders match.
Checks
Known limits
Not for merging yet: waiting for the owner's click-through.
🤖 Generated with Claude Code