Voice to Text
Turn speech into text — upload an audio file or record your voice, choose the language, and get a transcript. Your audio never leaves your device.
How to use the Voice to Text
- Pick the spoken language, and choose Upload or Record.
- Upload an audio file, or record straight from your microphone.
- Click Transcribe, then copy or download the text.
About the Voice to Text
Voice to Text turns spoken audio into written text, right in your browser. Upload a recording — a voice memo, an interview, a lecture, a meeting — or record straight from your microphone, choose the language being spoken, and get a transcript you can copy or download. It runs a Whisper speech-recognition model with WebAssembly on your own device, so your audio is never uploaded: it is private, free, and has no watermark or sign-up.
Picking the right language in the selector makes the result noticeably more accurate than leaving it on auto-detect, especially for shorter clips or accented speech — so set it before you transcribe. The first time you use the tool your browser downloads the AI model once (about 75 MB) and then caches it, so later runs start quickly. Because the work happens on your device rather than a server, it is best for clips up to a few minutes; very long recordings are slower and depend on your computer. Clear audio with little background noise transcribes best.
Frequently asked questions
Is my audio uploaded anywhere?
No. The speech recognition runs entirely in your browser with WebAssembly. Your recording never leaves your device — only the AI model is downloaded, once.
Which languages are supported?
The model is multilingual and handles dozens of languages. Choose the spoken language from the selector before transcribing for the most accurate result; leave it on Auto-detect if you are unsure.
Why is there a download the first time?
The first run downloads the speech-recognition model (about 75 MB) and caches it in your browser, so every run after that starts quickly.
What audio can I use, and how long?
Common formats like MP3, WAV, M4A, OGG and WebM work, plus live microphone recording. It is best for clips up to a few minutes — longer recordings still work but take more time, since everything runs on your own device.