Auto captions for YouTube
Generate an accurate SRT to upload alongside your video — better control than YouTube's auto-captions, and nothing leaves your device.
Free · No signup · Files never leave your device
Drag and drop a video or audio file
or click to browse
How it works
Drop your video or audio
Any MP4, MOV, MKV, MP3, or recording — it stays on your device, nothing is uploaded.
AI transcribes locally
OpenAI's Whisper speech-recognition model runs right in your browser via WebGPU or WebAssembly, with a timestamp for every line. Or upload your own SRT/VTT to skip this step.
Download your subtitles
Get a timestamped SRT or VTT file or a plain-text transcript — free, no watermark. Or burn the captions in / add a soft track instead.
Accurate captions, ready to upload
Captions widen your audience — viewers watching muted, non-native speakers, and anyone using a screen reader — and they give YouTube more text to index. This tool transcribes your video locally with Whisper and hands you a timestamped SRT you can proofread and then upload to YouTube Studio as a caption track.
Because it runs in your browser, there's no upload step and no queue: drop the file, review the transcript, download the SRT. If you'd rather have the captions baked into the picture for Shorts, switch to burn-in mode — the same transcript, painted into the frames.
Frequently asked questions
YouTube's auto-captions are decent but you can't easily correct them before publishing, and they're unavailable until processing finishes. Generating an accurate SRT here first lets you review and fix the text, then upload a polished caption track from day one — which also helps SEO and accessibility.
In YouTube Studio, open your video → Subtitles → Add language → Upload file → With timing, and pick the .srt this tool produced. Your captions appear as a selectable track viewers can turn on.
Yes. Upload an existing SRT or VTT file to skip transcription — then download it in another format, burn the captions into the video, or add them as a soft track.
No. The Whisper AI transcription runs locally in your browser via WebGPU or WebAssembly, and if you burn captions in, that re-encode runs on your device too. Your file never leaves your computer — once the page has loaded it even works offline.
Whisper was trained on about 100 languages, including English, Spanish, French, German, Chinese, Japanese, Arabic, and Hindi. Language is auto-detected by default; picking the spoken language can improve accuracy.
The Whisper speech model (~77–150 MB depending on your browser) downloads the first time you use the tool and is then cached. After that, transcription starts immediately.
Yes. The caption generator is completely free, with no signup, watermarks, or limits. We make money from our video translation service, not from this tool.