返回全部资讯
Rohan Paul (@rohanpaul_ai)T2·基座大模型

Voice Arena 发布真实通话语音识别基准 覆盖 5 种语言双人对话

原标题:Speech recognition evalualtion must test where voice agents actually work, which is phone calls betw...
62AI 研判分

Voice Arena 推出面向语音智能体真实场景的语音识别基准,首个数据集 Monsoon 聚焦日常手机通话中的双人对话,覆盖 5 种语言。该基准强调重叠说话、插话、短应答与电话线路压缩等真实难点,并公开排行榜与测试集。

报道正文

Rohan Paul (@rohanpaul_ai) · 2026-11-05 21

Speech recognition evalualtion must test where voice agents actually work, which is phone calls between two people.

Voice Arena’s new benchmark tests speech-to-text models on real calls made on everyday phones, in real rooms, with real background sound.

Two speakers make the job much harder for the model. People talk over each other, interrupt, drop in short “yeah” and “okay” replies, and the phone line compresses the audio. A support or sales agent deals with all of this every day.

↳ 引用 @shobhitbanga (Shobhit Banga): Drop 3/15: Introducing the Speech Recognition Benchmark on @voicearena_ai.

One benchmark for every use case of speech recognition, in every language. Tested on real conversations.

Dataset 1 is live: the Monsoon Conversational ASR Test Set, two-person conversations in 5 languages.

Leaderboard and public test sets at Voice Arena.

推文配图

推文配图