vikit.prompt.subtitle_extractor¶
Overview¶
A class to extract subtitles from a sound recording, |
Classes¶
- class vikit.prompt.subtitle_extractor.SubtitleExtractor¶
A class to extract subtitles from a sound recording, merge short subtitles into longer ones, or extract them as text tokens
Overview
Methods¶ build_subtitles_as_text_tokens(subtitles)Create blocks of subtitles
merge_short_subtitles(subtitles, min_duration)Merge subtitles which total duration is less than 7 seconds
Members
- build_subtitles_as_text_tokens(subtitles) list[str]¶
Create blocks of subtitles
- Parameters:
subtitles -- The subtitles to process
- Returns:
list of text tokens corresponding to the subtitles in some sort of human readable format
- merge_short_subtitles(subtitles, min_duration=7)¶
Merge subtitles which total duration is less than 7 seconds