fablecut_captions
Create timed caption clips from speech transcripts, style them as subtitle, karaoke, highlight, or pop, and export or import SRT and WebVTT subtitles.
Instructions
Captions from what is said: makes caption clips on the timeline from the clips' transcripts (transcribing first when there is none, like fablecut_transcribe), each one a text clip with wordTimes so its words follow the speech. Captions are cut at pauses and sentences, one or two lines as wide as the canvas allows. Styles: subtitle (plain, on a dark bar), karaoke (words light up as they are said), highlight (bold, the word being said in a colour — the reel look), pop (one to three big words at a time, each popping in). They go on the lane captions are already on, else the top video lane, else a new lane above it, and replace the captions there (replace:false keeps them). export:"srt"|"vtt" writes the timeline's captions to a subtitle file in exports/; import: a path to an SRT / WebVTT file (or its text) places its captions instead of the transcript's. remove:true deletes the captions. Captions are text clips: restyle them with updateClip, and make them after the cutting is done (a cut through a caption doesn't re-time its words — run this again). The same as the Transcript tab's Captions row.
Input Schema
| Name | Required | Description | Default |
|---|---|---|---|
| to | No | Only speech before this time | |
| from | No | Only speech from this time on (timeline seconds) | |
| size | No | Scale the style's text size, e.g. 1.2 (default 1) | |
| force | No | Also replace or remove locked captions (only when the user asked) | |
| lines | No | Lines per caption, 1–3 (default: the style's) | |
| model | No | Transcribing first: the Whisper model (see fablecut_transcribe) | |
| style | No | The look (default subtitle) | |
| track | No | The video lane to put them on (default: chosen as described) | |
| where | No | Transcribing first: where (see fablecut_transcribe) | |
| words | No | At most this many words per caption (pop and highlight have their own default) | |
| export | No | Write the timeline's captions to exports/ as SubRip or WebVTT, and change nothing | |
| import | No | An SRT or WebVTT file (absolute path, or a /media/… src) or its text: place its captions | |
| remove | No | Delete every caption clip | |
| clipIds | No | Caption only these clips' speech (default: every clip with sound) | |
| replace | No | Remove the captions already on the timeline first (default true) | |
| language | No | Transcribing first: the spoken language (default: detected) |