Transcribe audio & video

Billing is per minute of the file, and the duration is measured from the file itself. The per-minute price depends on the model you choose - the exact rate is shown on the page next to the model, and the total appears on the submit button.

Yes. Transcription runs on our GPU servers and is charged in credits, so you need to sign in and have credits on your balance.

Your audio or video is uploaded to our servers and the speech is turned into text there. Recordings and transcripts are deleted from our servers automatically after about six hours, so download and save the results somewhere permanent.

The larger models understand about a hundred languages. You can let the model detect the language automatically or pick one manually on the page before transcribing.

You can upload audio or video files. The price is calculated from the file duration (rounded up to whole minutes for billing). Accuracy is high on clean recordings; noise, a strong accent and several people talking at once bring it down.

Yes. The transcript comes with timestamps, and you can copy it, download it as a text file or export it as SRT subtitles ready for a video.