Burned subtitles.In your browser.
Drop a video. Style the captions. Export.
Drop a video to start.
or browse files.
MP4 · WEBM · MOV
Free. No sign-up. No watermark.
The whole editor runs on your device, for free.
Drop a video, transcribe, style, export. No sign-up, no upload, no watermark on the output.
A framework, not a style picker.
Every caption style is configuration, not a fixed preset. Tune colors, weight, animation, and layout, or extend the engine itself when you need something the library does not ship.
One video, many caption styles.
Group segments into sheets. Each sheet carries its own style and overrides, so an intro, a main cut and a tag line can read differently in the same video.
The little things are handled.
Auto-save. Language auto-detect. Speaker-based scene breaks. Pattern-based segment grouping. Filler punctuation cleanup. Gap-free joins. Per-word style overrides. More than fits in this little box.
How it works
From a dropped video to a finished clip.
- 01
Drop your video.
Drop the file. Whisper runs in your browser and the captions come back as text you can edit — nothing is uploaded.
- 02
Edit, structure, style.
Pick a caption style as a starting point. From there, every part is yours: typography, layout, animation, plus the controls that come with the template. Override per word or per scene where it matters. Use sheets when one video calls for several looks.
- 03
Export with captions inside.
A single MP4 with the captions rendered into every frame. Plays the same on TikTok, Reels, Shorts, or anywhere else.
The engine
A template framework, not a style picker.
Most subtitle tools give you a small grid of looks and a few color knobs. tscaps gives you the engine those looks are built on, with the full browser toolkit open.
Read the engine overview →- Anything a browser can paint
Templates can lean on whatever the browser already does. CSS gives them variable fonts, keyframed animations, and custom properties. SVG filters extend the toolkit when CSS is not enough.
- Templates are folders
A new style is a style.css and a small manifest. Drop the folder in and it shows up in the picker. No app rebuild, no plugin marketplace, no review queue.
- Open library, open contributions
Every built-in template is open source. Read the source, propose changes, or contribute new styles through the repo. The library belongs to whoever wants to extend it.
- Edit in the editor
The editor ships a code tab for the active sheet. Change the CSS, see the preview redraw. The same files live in the library when you want to push your work back.
Local or cloud
Cloud removes the limits.
Same editor underneath. Local runs on your device, so speed, accuracy, and mobile support are tied to your hardware. Cloud handles transcription on our servers and adds AI styling on top.
Free, private, on your device.
The full editor in your browser. No upload, no account, no watermark. The work runs on your hardware, with the limits that come with it.
- Free. No sign-up, no upload, no watermark.
- Transcription via in-browser Whisper.
- Bigger models are more accurate but slower and heavier on RAM. None match cloud.
- Mobile may present performance issues for transcription.
- Templates lose AI-context features (smart emphasis, contextual emojis, scene-aware styling).
- Open-source and self-hostable.
Faster, more accurate, smarter.
Server-side transcription and AI styling. Same editor underneath, but without the hardware limits.
- Fast transcription, regardless of your device.
- Higher accuracy with multi-speaker recognition.
- Works on mobile out of the box.
- AI styles your video per scene, per speaker, per word.
- AI picks where each caption breaks.
- Templates with contextual emphasis and emoji placement.
- Projects sync across devices.
Open core
The editor is open source.
tscaps is built on an open-source engine and a self-hostable web editor. The hosted version adds the parts that need a server: accounts, cloud transcription, project sync, AI assist. Everything else is the same code.
FAQ
Frequently asked questions.
Where does my video go?
It stays on your device. The editor opens the file locally, transcription runs in your browser via Whisper, and the export composites frames in-page. Nothing is uploaded.
What does it mean to burn the subtitles?
The captions are baked into the pixels of the exported video. There is no separate text track that can be toggled off or lost. The output plays the same way on TikTok, Reels, Shorts, or any other player.
What languages does the transcription handle?
The picker has auto-detect plus a curated list of common languages (English, Spanish, Portuguese, French, German, Italian, Dutch, Russian, Japanese, Chinese, Korean, Arabic, Hindi). The underlying engine handles many more behind auto-detect.
How does in-browser transcription compare to cloud?
In-browser transcription runs Whisper locally on your CPU or GPU, with model sizes capped so the page stays responsive. Cloud transcription uses a higher-accuracy model on our servers and adds AI editing features — hook detection, semantic tags, emoji suggestions. For short clips, offline work, or sensitive content, in-browser is the better fit; for accuracy on long videos and AI assistance, the cloud version is the right pick.
Can different parts of one video carry different caption styles?
Yes, that is what sheets are for. Group scenes into sheets and pick a caption style per sheet. An intro, a main cut and a tag line can all read differently in the same project.
Can I customize a caption style beyond the built-in knobs?
Yes. Each template ships its own controls, so most projects never need to leave the sliders. When you need to go further, the editor has a code tab that opens the active sheet's CSS and SVG filters for direct edits, with a preview that updates as you type.
Can I edit the transcript after it generates?
Yes. Rewrite words, retime per word or per scene, split lines, insert words mid-scene. You can also assign scenes to sheets by hand or auto-assign by pattern.
Can I create my own caption styles and share them?
Yes. A template is a folder with a small manifest and a CSS file. Build it locally, open a pull request, and once it lands in the repo it joins the built-in library.
Why is the local version free?
The editor and its engine are open source. The local editing path runs entirely on your machine, so it costs us nothing to operate. The cloud version exists for users who want hosted transcription, AI assistance, and project sync — that is what Pro subscriptions pay for.