Rohan Paul· @rohanpaul_ai · X·· 2026-09-24AI 评分46
AI 导读
VoiceArena 推出 Diarization Bench，用 139 段真实对话和统一评分规则对比 12 个说话人分离系统，NVIDIA Nemotron 3 以 14.72% DER 领先。该基准还揭示 collar 效应：同一输出在 250 ms collar 下 DER 为 4.29%，换成 0 ms 则升至 14.72%，并暴露重叠语音和说话人增多时准确率的快速下降。
正文 · 原文
VoiceArena launched Diarization Bench to compare 12 diarization systems on the same real-world audio and scoring rules.
It exposes how quickly accuracy drops with overlap and more speakers.
STT, TTS, conversational agents, and now diarization, all on @voicearena_ai
The number that got me is the collar effect: the collar is simply the timing tolerance around a speaker change.
NVIDIA Nemotron 3 scores 4.29% DER with a 250 ms collar. Score the exact same output at 0 ms and it becomes 14.72%.
Introducing Diarization Bench, @voicearena_ai's benchmark for who spoke when. The metric we measure is diarization error rate, or DER. It measures how much of a conversation a model attaches to the wrong person. Every second it misses, invents, or gives to another speaker counts against it. Here's what we did. 139 conversations recorded for this board, in rooms and on calls, with people interrupting each other throughout. 12 systems scored under a single goal: testing models in the real-world conditions they are actually deployed in. @nvidia's Nemotron 3 leads at 14.72%. 🧵在 X 查看被引用的帖子
来源：Rohan Paul · x.com