VoiceArena 推出 Diarization Bench，在相同的真实音频和评分规则下对比 12 个说话人分离系统。 它揭示了当语音重叠和说话人增多时，准确率下降得有多快。 STT、TTS、对话式智能体，如今再加上说话人分离，全都在 @voicearena_ai 上 让我震惊的是那个 collar 效应：collar 其实就是说话人切换时的时间容差。 NVIDIA Nemotron 3 在 250 ms collar 下得分为 4.29% DER。把完全相同的输出按 0 ms 评分，就变成 14.72%。
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%.
来源：@rohanpaul_ai · x.com