返回
原文
返回
AI 评分 45/100在 X 查看原推
导出 Markdown
Rohan Paul · @rohanpaul_ai · X·2026-09-24 01:06·1小时前
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。

Rohan Paul@rohanpaul_ai
45AI 编辑部评分，满分 100
2026-09-24 01:06· 1小时前
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。

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%.

Shobhit BangaIntroducing 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...

来源：Rohan Paul· x.com

导出 Markdown