Tools › Audio
Transcribe audio to text
Transcribe audio to text free: interviews, lectures, podcasts and voice memos in MP3, M4A, WAV or video, with timestamps. Whisper runs in-browser. TXT, SRT.
The speech model downloads once to your device; your recording never leaves it. Nothing starts until you choose a file.
Turn a recording into a draft you can edit
Whisper recognises speech and breaks it into editable lines with timestamps. Those times are useful for returning to a quote during an edit or checking a word you are unsure of. Transcription services usually need an uploaded recording. This tool decodes and analyses your audio in the browser on your device.
How it works
- Choose your recording and set the spoken language, or let Whisper detect it.
- Pick Fast for a first pass or Accurate for material you expect to be harder to hear, then start transcription.
- Play back unclear passages and edit the draft. Check names and numbers against the recording.
- Copy the result or download TXT. Keep timestamps if you plan to use them to find passages again.
Build notes from an interview or lecture
For an interview, keep timestamps during the first review so you can play the source when a name or figure looks wrong. Add speaker names yourself before sharing quoted passages. For a lecture, use the draft to find when a topic comes up, then turn the relevant passages into notes. For a voice memo with a single speaker, a plain text export is often easier to paste into a document.
Source files and transcript exports
Start with MP3, M4A, WAV or OGG, or open a video to transcribe its soundtrack. Use TXT with or without timestamps, or save timed lines as SRT or VTT. Free allows files up to 3 GB each. Pro lifts that cap on the device where you are signed in. Recordings with people talking over one another still need careful review.
What happens to the interview file
The network is used to fetch the speech model on first use. Your browser runs it against the audio locally, and the recording and resulting transcript stay off Nibbo’s server. Save the text before closing the page.
Questions
How can I transcribe audio to text for free?
Drop the recording on the Nibbo transcriber: MP3, M4A, WAV, OGG or a video. OpenAI’s Whisper model runs in your browser and writes the text with timestamps in any of 99 languages. It is free, needs no account and nothing is uploaded.
How do I get subtitles instead of plain text?
Use auto subtitles to download SRT or VTT, then burn them into the video if you need them in the picture.
How do I get a better transcript from a noisy recording?
Clean it first with Clean up a voice recording, which removes hiss, hum and room noise, then choose the Accurate model.
Which languages work?
All 99 languages Whisper knows, from English, Spanish and Hindi to Japanese, Arabic and Turkish. The language is detected automatically, or you can choose it.
Is it really private?
Yes. The model is downloaded once and runs in this tab. Nothing about your recording is sent anywhere.
Can it tell who is speaking?
Not yet. You get the words with timestamps; speaker labels are not added.
How accurate is it?
Clear speech comes out close to word for word. Background music, crosstalk and heavy accents cause more mistakes; the Accurate model helps.