speech-to-text

English and French speech recognition running in your browser.
Rust → WebAssembly + WebGPU. No server, no API key.
Your audio never leaves your machine.

Mobile mode: using a smaller, non-streaming model. Record your speech, then wait for transcription. For real-time streaming transcription, use a desktop browser with WebGPU.

Quantized Kyutai STT model (Mimi audio codec + text decoder), custom Rust inference compiled to WebAssembly, GPU acceleration via WebGPU.

Whisper Tiny Q5 via whisper.cpp compiled to WebAssembly. Record then transcribe (no streaming).

← trucs.ai