Skip to content

Fix the incomplete subtitles bug#13

Open
iceberg53 wants to merge 4 commits intoTejasQ:mainfrom
iceberg53:incomplete-transcript-fix
Open

Fix the incomplete subtitles bug#13
iceberg53 wants to merge 4 commits intoTejasQ:mainfrom
iceberg53:incomplete-transcript-fix

Conversation

@iceberg53
Copy link

This is a fix for the issue about the generated subtitles not covering the entire video .
It introduces an instruction for flushing the last part of the data stream.

Add explicitly the shell command to run for installing gen-subs to make it easier for
those new to the nodejs ecosystem.
This commit mainly replaces the use of custom string manipulation by
functions from the path module from node ( path.basename(),
path.resolve(), path.extname(), ...) .
It's done to account for Windows OS file paths
("C:\\folder\\subfolder\\file.ext") which are different from POSIX ones
("/home/user/folder/file.ext") .
That difference led to file names with character sequences that caused
an error when calling ffmpeg.
This commit introduces an instruction for the vosk recognizer to
transcribe the last part of the audio file.
Before this change, the last part of the audio wouldn't be converted to
text resulting in incomplete subtitles.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant