Text Tools
Audio & Video Transcriber
🔒 Runs in your browser
Transcribe any audio or video file to text, SRT, or VTT — entirely in your browser. No upload, no account, no API key. Your file never leaves your device.
How to use this tool
- Select an audio or video file from your device.
- On first use the Whisper speech model downloads once and your browser caches it; after that, transcription runs on-device.
- Read the transcript and export it as plain text, SRT, or VTT.
Drop an audio or video file here or click to upload
MP3, WAV, M4A, MP4, WebM and more — transcribed in your browser
Transcription runs entirely in your browser via a self-hosted speech model (ONNX/WASM, WebGPU-accelerated when available). Your audio is never uploaded. The model downloads once on first use, then your browser caches it.
Frequently Asked Questions
- How does in-browser transcription work?
- It runs an OpenAI Whisper speech-recognition model in your browser via ONNX/WASM, accelerated by WebGPU when your device supports it.
- What gets sent over the network?
- Only the model file, which downloads once on first use (it is sizeable, often tens to a couple hundred MB) and is then cached. After that the model is reused offline — your audio or video is decoded and transcribed on-device and is never uploaded.
- What can I export?
- Plain text for documents, or SRT and VTT subtitle files with timing for video.
- How accurate is it and what are the limits?
- Quality is good on clear speech but drops with heavy noise, crosstalk, or strong accents, and long files take time and depend on your device's memory and speed.
Processing EU personal data in your workflow?
GDPR compliance is required. EuroComply maps your obligations in minutes.