Skip to content
Utilboxes

Voice to Text Converter

Transcribe a voice note or a short recording into text. The speech recognition model runs on your device, so the audio is never uploaded.

Files never leave your device

How to use the Voice to Text Converter

  1. 1Add an audio file — a WhatsApp or Telegram voice note, a phone recording, or an MP3, M4A, WAV, OGG or WebM clip.
  2. 2Choose the language, or leave it on automatic, and pick the faster or more accurate model.
  3. 3Press Transcribe. The first time, the model downloads; after that it is cached.
  4. 4Copy the text or download it — with timestamps and subtitles if you turned them on.

How it works

Transcription uses Whisper, OpenAI's open-source speech recognition model, converted to run in the browser with Transformers.js and ONNX Runtime Web. Your browser first decodes the audio file and resamples it to 16 kHz mono, the format Whisper expects; the model then turns 30-second windows of sound into text, overlapping them slightly so words at the edges are not lost.

The model files are the only thing downloaded — about 40 MB for the faster model and about 80 MB for the more accurate one — fetched once from Hugging Face and then cached by the browser. The audio itself stays on your device and is not sent to this site, OpenAI or anyone else, which matters for private voice messages.

Recognition runs on your own processor in a background thread, so a one-minute clip can take a few seconds on a recent laptop and a minute or more on an older phone. Clips up to 10 minutes are accepted.

Whisper is accurate on clear speech but can mishear names and specialist terms, struggles when people talk over each other, and occasionally invents words during long silences. Read the transcript before relying on it.

Frequently asked questions

Can I record straight from my microphone?
Not at the moment. The tool transcribes audio files, which covers voice notes saved from messaging apps and recordings from your phone's voice memo app.
Which audio formats work?
Anything your browser can decode: MP3, M4A and AAC, WAV, OGG and Opus voice notes, WebM and FLAC. If a file will not decode in one browser, try another, such as Chrome or Firefox, or convert it to MP3.
How do I save a WhatsApp voice message as a file?
On a phone, press and hold the voice message, choose Share, and save it to your files. On WhatsApp for desktop or the web, open the message menu and choose Download.
Is the audio really kept private?
Yes. Only the model is downloaded; the recording is decoded and transcribed on your device. You can confirm in your browser's developer tools that no request carries the audio.

Related tools

Image to Text (OCR)Read the words out of a screenshot or photo of a document.Image ToolsWord CounterLive word, character, sentence and reading-time counts.Text CleanerFix smart quotes, stray spaces and broken line breaks.Character CounterCount characters against Twitter, SMS and meta-tag limits.