Add subtitles to your video
word by word, automatically

Whisper handles around a hundred, and detects which one it is hearing on its own - you do not pick it beforehand. Accents and code-switching are handled well; very noisy audio is where any speech model starts to guess.

The transcription step takes files up to 25 MB, which is a few minutes of a normal clip. Longer than that, cut it into parts - and the editor lets you paste the pieces back into one timeline.

Yes, all of it: the words themselves, where the lines break, how many words show at once, the position, the colours, which words are emphasised. The generation is a starting point, not a locked result.

No. The subtitles are rendered into the video itself, which is what short-form platforms need - a separate subtitle file is ignored by TikTok, Reels and Shorts feeds.

Yes. Music-only passages are recognised as such and left without text rather than transcribed into nonsense, and the speech around them keeps its timing.

Creating and subtitling a video is free; only the final render consumes credits, and a free account gets renders to try it with. You see the cost before you start it.