Speech to Text
This runs Whisper tiny, an OpenAI speech-recognition model compiled to WebAssembly (WASM) and executed in your browser. The first run downloads roughly 40 MB of model data (cached afterwards); transcription is offline and your audio is never uploaded.
What is Whisper / WASM?
Whisper is an automatic speech-recognition model trained on multilingual audio. WebAssembly (WASM) lets it run at near-native speed inside the browser tab. The tiny variant is fast and private, but less accurate on accents, noise and technical vocabulary than larger models.
English only in this build. Long files are processed in 30-second windows and may take a while on a laptop.
This tool runs entirely in your browser. Your inputs and files are processed locally and are never sent to or stored on Kinetikatech servers.
Related pages
Server problem? Hacked site? Slow website?
Critical issues answered within 30 minutes, 24 hours a day, 7 days a week.