Pulse 在 Voice Arena 说话人识别与转录榜单排名第一

返回 Rohan Paul

Rohan Paul· @rohanpaul_ai · X· 2026-10-07 12:19 · 原发布时间 AI 评分34

自动核验发布 · 本文由系统生成并完成证据核验，未经人工审稿。

AI 导读

@smallest_AI 的 Pulse 在 Voice Arena 的说话人识别与转录赛道中，以 24.4% 的说话人识别错误率位列七个系统之首；第二名 ElevenLabs Scribe v2 为 40.7%。该测试使用真实对话的远场录音作为输入，并依据每位说话者的独立麦克风录音制作标注。

正文 · 原文

该语言的正文暂不可用，当前显示已有版本。

Congrats @smallest_AI on taking #1 for diarization + ASR on @voicearena_ai.

Its a very effective real-world kind of benchmark: > models hear far-field room audio (i.e. microphone is some distance away from the people speaking), but they’re scored against labels built from a mic on every speaker.

> Realistic input, accurate ground truth. That combination is rare.

Many diarization benchmarks either use clean audio or have messy labels. @voicearena_ai avoids both: models get far-field recordings of real in-person conversations, and the ground truth comes from individual mics on each speaker. #1 really means something.

Congrats @smallest_AI.

引用Voice Arena@voicearena_ai

Pulse by @smallest_AI currently ranks #1 in the Diarization + ASR track by DER, as the newest entry on the Voice Arena Diarization Bench. The Diarization + ASR track ranks Speaker-Attributed ASR systems that return speaker labels together with the transcript. Seven systems are in it today. Pulse scores 24.4% DER at a strict 0 ms collar. The next system, ElevenLabs Scribe v2, scores 40.7%. That is 1.7x lower error than the next best in the track. Most of the gap is missed speech. For every other system in the track, missed speech is the largest part of the error, above 30% for each of them. Pulse misses 4.4% of reference speech, the lowest in the track. On the full board of 13 systems, dedicated diarization models and Speaker-Attributed ASR systems together, Pulse places 5th. It is the highest-ranked Speaker-Attributed ASR system on the board and sits ahead of two dedicated diarization models. The result holds across settings: In-person recordings : 24.7% DER In online calls : 22.4% DER Congratulations to @smallest_AI 👏 View full results, per track and per condition: https://t.co/T9H6hLEUxR

在 X 查看被引用的帖子

发现内容有误？提交纠错