VibeVoice-ASR-BitNet
Microsoft🇺🇸 USMajor lab· Jul 24, 2026
Audioopen weights
Compressed speech recognition model derived from VibeVoice-ASR for real-time transcription on CPUs without a GPU. The model card reports RTF below 1 from 3 threads at a total size of 1.58 GB.
Summary of the model card
- License
- mit
- Modalities
- audio → text
1 variant
| Standard | HF |
Sources
HuggingFacePaperDeutsch
Komprimierte Spracherkennungs-Variante von VibeVoice-ASR für Echtzeit-Transkription auf CPUs ohne GPU. Laut Modelcard liegt der RTF ab 3 Threads unter 1 bei einer Gesamtgröße von 1.58 GB.
Provenance
- Detected
- Sep 27, 2026
- First source
- huggingface
- Description
- Summary of the model card
- Editorially reviewed
- —