How it works
- Pick a model above: Whisper for 99+ languages, or Parakeet for fast English.
- Load the model. The first time, it downloads into your browser (about 565 MB for Whisper, 670 MB for Parakeet) and is reused after that.
- Upload an audio file (MP3, WAV, WebM or another format your browser can play) or record from your microphone.
- Copy the transcript when it's done.
How is this different from other free transcription sites?
Most free transcription tools upload your audio to their servers. Here the Whisper and Parakeet models run inside your browser, so the recording is processed on your own computer.
That makes it a good fit for interviews, voice memos, lectures and anything you'd rather not hand to a third party.
Which model should I use?
Whisper Large V3 Turbo handles 99+ languages and accents. Parakeet TDT 0.6B is English only and built for speed.
Both download once and then load from your browser's cache, so the second transcription starts much faster than the first.
What do I need?
A recent desktop browser and enough free memory for the model. A newer laptop transcribes faster than an older one, because the work happens on your hardware.
Frequently asked questions
Is audio to text transcription free?
Yes. The in-browser Whisper and Parakeet models are free to use with no account.
Is my audio uploaded to a server?
Not when you use the Whisper or Parakeet models. They run in your browser, so the audio is processed on your device.
What audio formats work?
Common audio formats including MP3, WAV, WebM and OGG. If a file won't open, convert it to MP3 first.
What languages are supported?
Whisper supports 99+ languages. Parakeet is English only.
Can I transcribe on my iPhone?
For on-device transcription on iPhone, our DictaWiz app records, transcribes and summarizes entirely on the phone.