Source
VibeVoice-ASR official documentation
Read the primary source.
Reference illustration, not a screenshot of this source.
VibeVoice-ASR
Speech-recognition model · Voice
Model with documented structured transcription, speaker and timing output.
Use it for
Evaluate when who-spoke-when output matters.
Choose something else for
Not the same deployment as the CPU BitNet runtime already used locally.
Give it
Long audio recording
Get back
Structured transcript; Speaker labels; Timestamps
How it is operated
Model/runtime deployment
Availability
Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.
How charging works
Compute and model/weight terms; hosted wrappers charge separately.
Where processing happens
Self-hosted if deployed locally, otherwise provider-dependent.
Language support
Exact model version determines support.
Speaker-labelled transcription
★★★☆☆ 3/5 for this task
Good capability fit on paper, but the required output needs verification in the chosen runtime before becoming the default.
Decision: Evaluate when who-spoke-when output matters.
Limits and things to check
- Do not advertise structured output as proven locally. Streaming ASR is a separate release.
Not the same deployment as the CPU BitNet runtime already used locally. Do not advertise structured output as proven locally. Streaming ASR is a separate release.
Evidence
Official capability documentation; no listening benchmark
Catalogue review: 2026-09-05. The evidence note identifies what was verified. This date does not imply a fresh tool test or verification of every linked source.
