Speech to Text
Transcribe audio and speech to text — free, private, runs entirely in your browser. No files are uploaded.
Transcribe any audio file to text instantly using AI speech recognition. Your files never leave your device — all processing happens locally in your browser using WebAssembly. No sign-up, no limits, no data collection.
Choose your spoken language and quality, then download the model. It runs entirely in your browser and is cached for next time.
High quality is more accurate and uses a larger one-time download. Normal works well for clear speech; for some languages such as Chinese, High is noticeably better.
Drop an audio file here or click to browse
Supports MP3, WAV, M4A, OGG, FLAC, WebM, MP4
Transcript
No data transfer — all data stays on your device!
How Speech to Text Works
WebConverter uses whisper.cpp compiled to WebAssembly with SIMD optimizations. Pick your spoken language and a quality tier (Normal is about 24 MB, High is about 41 MB), then download the model once. It is cached in your browser for future visits. Audio is decoded locally, processed by the speech recognition engine, and the transcript is shown with timestamps. Everything runs on your device, and nothing is uploaded to any server.
Features
Near-Zero CO₂
Server-based converters upload your file, process it remotely, and send it back — consuming network and data-centre energy on every request. WebConverter runs entirely on your device, so the only energy used is your local CPU. No data transfer, no server electricity.
Install for Offline Use
WebConverter works offline as a Progressive Web App. Click "Add to Home Screen" on mobile or the install button in your browser's address bar. Once installed, you can convert files without any internet connection.
