Transcribe Audio to Text
Turn the speech in a recording into text: interviews, lectures, meetings, voice notes and videos. The language is detected by itself, and the result is plain text or timed subtitles (SRT or VTT). It is free, needs no account, and adds no watermark to your result.
Transcribe Audio to Text
Add your files. Recommended settings are ready, so you can process in one click.
Accepts MP3, WAV, M4A…, MP4, MOV, WebM…
Up to 100 files · 300 MB · No account- 1Add files
- 2Choose an action
- 3Download or keep going
What people use it for.
What happens to your file.
The sound is taken from the file and the speech is found and written out by Whisper, an open speech recognition model (MIT licence), running on this site's own server. Pauses of two seconds or more start a new paragraph. A long file runs in the background with a progress bar.
Where it stops: Up to an hour of audio per file. Accuracy depends on the recording: clear speech close to the microphone comes out best, while strong background music, crosstalk and heavy accents produce more mistakes. Speakers are not labelled by name.
Everything you need, without a paywall.
Common questions.
Which languages are recognised?
Around 100, including English, Arabic, Spanish, Indonesian, Turkish, Vietnamese, French, German, Hindi, Urdu, Chinese and Japanese. Choose Detect automatically or pick the language.
Is my recording sent to another company?
No. The speech recognition runs on this site's server, and the file is deleted when the job ends.
You might also need.
Make subtitles from the speech in a video automatically: an SRT or VTT file timed to the words, or the video with them burned in.
Extract the audio track from a video and save it as MP3, WAV, M4A or FLAC.
Lower hiss, hum and fan noise in a recording while the voice or music stays clear.
Translate a Word, PDF, text, Markdown or subtitle file between 16 languages while its layout stays the same.