๐๏ธ VibeVoice-ASR-BitNet
Multilingual speech recognition with a ternary (1.58-bit) BitNet decoder, running entirely on CPU through the authors' ggml engine.
Supports English, Chinese, French, Italian, Korean, Portuguese and Vietnamese.
Examples (LibriSpeech ยท FLEURS fr ยท FLEURS zh)
Model: microsoft/VibeVoice-ASR-BitNet ยท Engine: VibeASR.cpp ยท Paper: VibeVoice-ASR-BitNet Technical Report