voixful-nemotron-speech-streaming-en-0.6b

Apple Core AI (Palette4-quantized) conversion of nvidia/nemotron-speech-streaming-en-0.6b — a cache-aware streaming RNN-T for lowest-latency on-device English dictation on Apple Silicon. Used by the Voixful engine and PrivoVoice.

License & attribution

Weights derive from nvidia/nemotron-speech-streaming-en-0.6b, © NVIDIA, under the NVIDIA Open Model License, and are redistributed under those terms with attribution. The Voixful conversion pipeline is Apache-2.0.