What formats work
Upload MP4, MOV, or WebM and you'll get back a captioned 9:16 MP4, encoded on your own device. We don't offer an SRT/VTT download, and there's no cloud queue sitting between you and the file.
Upload your talking-head clip and tell us what language was spoken. We transcribe it, you go through and fix whatever it got wrong, then burn a 9:16 MP4 right in your browser. Translation is there if you need a second language on screen — otherwise it stays off.
MP4, MOV, or WebM, up to 500 MB. Choose the language spoken in the video from the list below so the transcript starts off right.
Each caption lands as the same time range the player shows you. Fix a misheard word, highlight one for emphasis, set the color and font.
Roughly 10–18 MB per minute at 720p, encoded right in your browser. Keep the tab open until the download finishes.
There are 91Deepgram listen locales to pick from, so you tell us what's actually being spoken instead of hoping auto-detect gets it. Translate is a separate switch — leave it off and captions stay in that language, or turn it on to put a second language on screen.
Upload MP4, MOV, or WebM and you'll get back a captioned 9:16 MP4, encoded on your own device. We don't offer an SRT/VTT download, and there's no cloud queue sitting between you and the file.
A minute counts against your workspace once we've transcribed the upload — not before. Free plans start with 50 minutes. See the full pricing breakdown.
You get 50 free transcription minutes to try it on a real clip. Create a workspace when you are ready.