AIHacker Newsabout 2 hours ago
Whistle Model Released for Mobile Devices
5 min read
Developers release Whistle, a 16.9 MB speech recognition model running on device CPUs without dependencies. It transcribes 30 seconds of audio in 5.9 ms at 5 seconds and supports seven languages including English and German. The engine ships prebuilt for seventeen targets from macOS to WASI components.
Level
Hype check
Tap to vote and see what everyone thinks.