Summary
Whistle is Cactus's compact speech-recognition model for phones, wearables, and other CPU-based devices. A 16.9 MB quantized model runs in the same C++ engine as Needle and can load beside it for a speech-to-tool-call pipeline.
Whistle is Cactus's compact speech-recognition model for phones, wearables, and other CPU-based devices. A 16.9 MB quantized model runs in the same C++ engine as Needle and can load beside it for a speech-to-tool-call pipeline.
Developers can build private voice interfaces that transcribe locally rather than depending on remote audio processing. Core capabilities include 16.9 MB model download, CPU inference, Seven supported languages, and Live transcription.
Public code or model weights are linked from the project, with setup instructions for running or adapting the system. Additional characteristics include Whole-clip transcription, Keyword conditioning, Needle engine compatibility, and On-device browser demonstration.
Get more likes & reach the top of search results by adding this button on your site!


Get top updates in AI to your inbox every weekend. It's free!