Event ·Multi-source

Nvidia 开源 1 亿参数说话人分离模型 Nemotron 3 Diarization

Nvidia 发布 Nemotron 3 Diarization，参数量约 1 亿，可实时判断一段对话中正在说话的人是谁。模型最多支持同时识别 8 个说话人，并免费开放使用。这类说话人分离任务常用于会议转写、字幕生成等场景。

Current read

Nvidia 免费放了个 1 亿参数的小模型，能实时分辨最多 8 个说话人，做会议转写或字幕的可以试试。

11 sources67° AI heatLast updated 2026/9/27 11:01:13

Evidence

11 sources

Primary sourceDecoder

2026/9/27 11:01:13View source

Related source 1OpenBMB

2026/9/28 13:00:03View source

Related source 2Aravind Srinivas

2026/9/28 15:19:08View source

Related source 3Yuchen Jin

2026/9/29 03:12:14View source

Related source 4Genspark

2026/9/25 17:33:57View source

Related source 5shao__meng

2026/9/28 02:45:06View source

Related source 6IT之家

2026/9/28 09:33:59View source

Related source 7Tech in Asia

2026/9/28 09:47:12View source

Related source 8rohanpaul_ai

2026/9/28 10:41:50View source

Related source 9Artificial Analysis

2026/9/28 12:28:49View source

Related source 10Thomas Wolf

2026/9/28 13:52:55View source

Conflict check

Dedup data only shows that these sources discuss the same event, not that they agree. There is no structured supporting or refuting evidence yet, so no conflict is inferred automatically.

Open the story