Model Detail

VibeVoice-ASR-4bit is an audio model with 4B parameters released by mlx-community. The model is registered under the automatic-speech-recognition pipeline tag on Hugging Face, distributed under the permissive mit license.

VibeVoice-ASR-4bit ships with 4B parameters. Total weight footprint is approximately 8.3 GB, which is the relevant figure when planning local-inference VRAM. The mit license is permissive, allowing commercial deployment and derivative work without per-seat fees, though attribution requirements still apply.

VibeVoice-ASR-4bit is best fit for speech recognition, transcription, or speech synthesis depending on the task head. Treat this as a starting matrix rather than a benchmark verdict — the right deployment usually depends on the specific evaluation suite that mirrors your workload.