Generate subtitles from a video automatically: drop the file and Whisper writes timestamped captions you can download as .srt or .vtt. It runs on your own device, so unreleased footage and private recordings stay private. Click any cue to hear it and fix a word before you export, and the subtitle file loads straight into video players, editors and YouTube uploads.
How it works
- Drop the video, or an audio file. The audio is decoded on your device.
- Pick the spoken language and a model, Fast or Accurate. This page opens with subtitle output already selected.
- Transcribe, click any cue to check it against the audio, then download .srt or .vtt.
Questions
- Should I download SRT or VTT?
- SRT for most players, editors and YouTube; it is the most widely supported subtitle format. VTT (WebVTT) is what HTML5 video on a web page expects. The cues and timings are the same in both, so pick whichever your player asks for.
- Can I burn the subtitles into the video?
- Not on this page, which produces the subtitle file. Take the .srt to the burn-in tool and the captions are drawn into the frames of the video, so they show on platforms that do not accept subtitle files.
- Does my file get uploaded?
- No. The models and all processing run inside your browser on your own device. You can verify this yourself: open your browser’s DevTools, watch the Network tab, and process a file: there are no uploads. After the first visit most tools also work fully offline.
- How accurate is it?
- The models are OpenAI’s Whisper (tiny and base tiers). On clear speech they are very usable; heavy accents, crosstalk and noisy rooms will show errors. The Accurate tier is noticeably better on difficult audio.
- Which languages work?
- Fifteen, picked in the Language setting: English, Spanish, French, German, Italian, Portuguese, Dutch, Russian, Polish, Turkish, Japanese, Korean, Chinese, Arabic and Hindi. The decoder is conditioned on the one you pick, so it transcribes that language rather than approximating it in English — which also means picking the wrong one returns confident nonsense. The weights understand 99 languages; these are the ones we measured as good enough to offer.
- How long does it take?
- Transcription runs on your processor, so speed depends on the machine: expect around a quarter of the recording’s length, and less on the Fast model than on Accurate.