Audio to Text
Turn a recording, an interview or a voice note into text you can copy, edit and search, without sending the audio anywhere.
or drop them here · up to 50 MB per file, free
Bigger files? Pro converts files up to 500 MBOver the free limits: you can unlock the download from €1.99.
Working on it…
Your file is ready
Free · No sign-up · No watermark · Your files never leave your device
Audio to Text in 3 steps
- Select your audio files (MP3, M4A, WAV, OGG, Opus and more), or drop them on the box.
- Leave the language on Detect automatically, or choose it, and click Transcribe.
- Correct the text if needed, then copy it, download it as TXT or get subtitles as SRT.
Speech recognition that runs on your own device
Listening to a recording and typing it out takes four to five times its length. This tool does it for you with Whisper, the speech recognition model published by OpenAI, which understands more than 40 languages and copes well with accents, phone microphones and background noise. Use it for interviews, lectures, meetings, podcasts, dictated notes or voice messages: the text appears in a box on the page with punctuation and paragraphs, ready to paste into a document, an email or a translator.
Most transcription sites upload your audio to their servers, ask you to sign up and stop after a few free minutes. Here the model is downloaded once from this site and runs inside your browser, so the recording never leaves your computer or phone and there is no limit on minutes. The first time it downloads about 80 MB (the Most accurate setting, about 250 MB); after that it starts straight away. With the Fast setting a minute of audio takes about 20 seconds on a recent computer, and longer on a phone. Clear speech close to the microphone gives the best text; music, people talking over each other and very noisy rooms give more mistakes.
Frequently asked questions
How do I transcribe audio to text for free?
Select the audio file here and click Transcribe. The text appears on the page after a short wait: copy it, correct it or download it as a TXT file. You can also download subtitles with timings (SRT). There is no account, no watermark and no limit on minutes.
Which languages can it transcribe?
More than 40, including English, Spanish, French, German, Italian, Portuguese, Dutch, Polish, Turkish, Japanese, Chinese, Korean, Russian and Arabic. The language is detected from the first seconds of the recording. If a recording starts with music or in another language, choose the language yourself for a better result.
How accurate is the transcription?
With clear speech it gets most words right, with punctuation. Accents, fast speech, background noise and several people talking at once cause more mistakes, so read the text before you use it: you can correct it in the box. The Most accurate setting uses a larger model that makes fewer mistakes but takes longer.
Is there a time limit?
No limit on minutes. Files up to 50 MB are free, which is several hours of a voice note or about 50 minutes of a 128 kbps MP3. Long recordings simply take longer: keep the tab open while the progress bar moves. Larger files are possible with Pro.
Is my audio uploaded to a server?
No. The transcription runs inside your browser, on your own device, so private conversations, interviews and medical or legal recordings are never sent to us or stored anywhere. Only the speech model is downloaded from this site, once, and kept in your browser for next time.
Related tools
Part of: Audio tools · WhatsApp file converter · iPhone file converter