model

Nemotron 3 Diarization

NVIDIA speaker-diarization model of about 100 million parameters that labels up to eight speakers in streaming or recorded audio, using arrival-ordered speaker channels and producing timestamps rather than text.

Known aliases

Nemotron-3-Diarization-preview
NVIDIA Nemotron 3 Diarization

Relationships

No evidence-backed relationships are recorded.

Current clusters

build2 publishers

An evaluation-only preview licence gates NVIDIA's top-ranked diarization model

NVIDIA reports a 14.72% diarization error rate for the 100M-parameter Nemotron 3 Diarization, first on Voice Arena's initial leaderboard. Its Hugging Face preview page confines the weights to internal testing and evaluation.

Publishers:huggingface.coruntimewire.com

Reality

Evidence54
Adoption16
Hype gap+32
Incentives78
Confidence63