Audio
Transcribe Audio to Text
Turn audio or video into text using AI, entirely in your browser. Your recording never leaves your device — unlike every cloud transcription service.
Drop your audio or video here
or click to browse · MP3, WAV, M4A, MP4, MOV and more
This transcription tool runs Whisper — an AI speech recognition model from OpenAI — directly in your browser using ONNX Runtime Web (WebAssembly). There is no server, no account, and no upload: the model itself is downloaded once and cached by your browser, and every transcription after that runs entirely offline on your machine. Nothing about your recording is ever sent over the network — you can confirm this yourself by watching the Network tab in your browser's developer tools while a transcription runs.
Works on audio files (MP3, WAV, M4A, AAC, OGG, FLAC) and video files (MP4, MOV, MKV, WEBM, AVI) — the audio track is extracted automatically. Choose the Base model for the most accurate results, or Tiny for a faster, smaller download. Turn on timestamps to export subtitle files (.srt/.vtt) alongside the plain transcript (.txt/.docx).
How to Use
- 1
Drop an audio or video file onto the uploader or click to browse.
- 2
Choose the Base (more accurate) or Tiny (faster) model, pick a language setting, and optionally turn on timestamps.
- 3
Click "Transcribe". On first use, the AI model is downloaded and cached by your browser — after that, it starts immediately. Progress is shown throughout.
- 4
Read the transcript on screen, copy it, or download as .txt, .docx, and — if timestamps were enabled — .srt or .vtt subtitles.
Frequently Asked Questions
Is my recording uploaded to a server?
No. Transcription runs entirely in your browser via ONNX Runtime Web. Your recording never leaves your device — verifiable in your browser's Network tab during a run.
Can I transcribe a video?
Yes — drop in a video file and its audio track is extracted and transcribed automatically. The video itself is never uploaded either.
How accurate is it?
Very good, but not perfect — accuracy varies with audio quality and accents. Clear speech transcribes very well; noisy audio or heavy accents will produce more errors, as with any speech-to-text system.
Base vs Tiny — which should I pick?
Base is more accurate and is the default. Tiny is smaller and faster — a reasonable choice on slower devices or for a quick draft transcript.
Does this work on mobile?
It can, but this is a heavy, desktop-oriented tool. On mobile or lower-memory devices you'll see a clear warning before starting, and very long files are blocked outright to avoid the tab crashing. For best results, use a desktop browser.