Subtitle engines
- Edge (recommended)
- Whisper
- Disabled
The Best for: Most use cases, fast generation, no extra setup.
edge engine uses timing data from Edge TTS voice synthesis to sync subtitles. It is fast, requires no model download, and works well for most videos.config.toml
Engine comparison
| Feature | edge | whisper |
|---|---|---|
| Speed | Fast | Slow |
| Accuracy | Good | Higher |
| Model download | None | ~3 GB |
| GPU support | No | Optional (cuda) |
| Recommended | Yes | When edge is insufficient |
Subtitle styling
These parameters control how subtitles appear in the video. Set them in the Web UI or pass them in the API request body.Vertical position:
"bottom", "top", or "center".Font file name. Font files must be present in the
resource/fonts/ directory. To add a custom font, place the .ttf or .ttc file there and use its filename here.Font size in pixels.
Subtitle text color as a hex string, e.g.
"#FFFFFF" for white.Background fill behind subtitle text. Set to
true for a default background, false to disable, or a hex color string for a custom color.Outline (stroke) color for subtitle text, e.g.
"#000000" for black.Width of the text outline in pixels.
Adding custom fonts
Place any.ttf or .ttc font file into the resource/fonts/ directory in the project root, then reference it by filename in font_name.