Add Captions to Video with Local Transcription
Velocaption’s captions editor is the finishing workspace for enriched videos. It transcribes your video locally with Whisper, then gives you the tools to refine the result: word-level timing, reusable text and caption styles, translations, overlays, layouts, background music, subtitles, and frame-accurate export.
How It Works
-
Import your video: drag a file into Velocaption, open a cleaned clip from Silence Remover, or bring in footage from Finder. The editor runs a quick probe and generates a proxy for smooth playback.
-
Transcribe locally: pick a Whisper model (tiny to large), set the language, and hit Transcribe. Word-level timestamps and confidence scores appear in the sidebar when transcription finishes.
-
Edit the transcript: double-click any word to fix it. Confidence underlines (yellow, orange, red) highlight where Whisper was uncertain so you know where to focus.
-
Style and position: pick a caption style from the library or build your own reusable style (font, color, background, shadow). Drag captions to reposition them per phrase. Add enter/exit animations.
-
Add finishing media: layer text titles, translations, images (logos, watermarks), sound effects, background music, and picture-in-picture video clips. Each overlay has its own timeline track for precise timing control.
-
Choose a layout: landscape (16:9) or portrait (9:16) with independent settings. Add a floating webcam bubble, set the background (solid, gradient, or image).
-
Export: Remotion renders every frame locally. Captions, overlays, animations, and layout are baked into the final .mp4. What you see in the preview is what you get in the export.
Key Details
Captioning runs locally. Whisper runs on your Mac after the model is available, so you do not need per-minute API keys for the core caption workflow.
Two rendering paths, one result. The live preview and the Remotion export use shared rendering code. Style a caption in the editor and the exported video matches exactly.
Word-level precision. Every word has its own timestamp, confidence score, and edit state. Split words, merge them, reposition phrases: all at the individual word level.
What’s Next
- Local Whisper Transcription: model selection, language options, and how the transcription pipeline works
- Transcript Review: low-confidence words, review order, and text size for long correction passes
- Caption Tracks: original and translated caption tracks for multi-language projects
- Edit Words in the Transcript: confidence underlines, editing, reverting, and the right-click context menu
- Caption Styling: fonts, colors, backgrounds, shadows, and the style library
- Export Captioned Video: resolution, quality, and the Remotion rendering process
Keep the edit on your Mac.
Choose a license to record, clean, caption, and export without moving private footage into a browser editor.