Upload an MP3 (a podcast episode, an interview, a dictaphone recording, a call) and Wave AI turns it into text with each speaker labelled and timestamped, plus a short summary. No account, no watermark.
Drop an MP3 file here
Any .mp3, up to 200MB (about 3 hours at 128 kbps). Free, no account needed.
MP3 is the format most podcasts, dictaphones and audio editors hand you by default. A few common places to grab one:
The limit is 200MB per file, so length depends on the bitrate. At 128 kbps that's about 3 hours; at 64 kbps (common for spoken-word podcasts) it's over 6. You can transcribe up to 5 files a day for free.
Not much for speech. Anything from 64 kbps up sounds the same to the recognizer. What matters more is how close the microphone was, background noise, and people talking over each other.
Yes. Every line of the transcript shows who spoke and when. You can copy it, download it as .txt, or download .srt and .vtt subtitle files when timing is available.
Yes. The transcript labels each voice (Speaker A, Speaker B…) so you can see who said what. Rename them in your own copy once you know who's who.
The file is uploaded to temporary storage and transcribed. We may then keep a private copy of the audio and transcript to improve Wave's speech recognition. It's never published. Don't upload recordings you don't want to share with us.
Related: Free Audio Transcription: Transcribe Audio to Text with AI · Turn Any Audio Into Searchable Text
Product
Use Cases
Compare
All rights reserved
Made with love in New York City