跳到正文
原文
HuggingFace Daily Papers(社区热门论文)· HuggingFace Daily Papers(社区热门论文)·· 8 天前AI 评分56

VoxParity 基准测试语音智能体对音频线索的响应能力

AI 导读

VoxParity 基准用 14 个行业 183 个场景测试语音智能体是否根据音频线索改变行动,同一文本的音频变化(如 mayday 呼救、儿童声音下注、恐惧低语)应触发不同的工具调用。

正文

View PDF HTML (experimental)

Abstract:A voice agent can handle almost every call on the words alone and still fail the few its sector's rules were written for. Emergency-call standards, fraud guidance, radio phraseology and vulnerability rules recognise that how a caller sounds, or what else is audible, can change the right action. VoxParity tests whether agents act on it. In 183 scenarios from 14 sectors, one transcript stays fixed while the audio changes (a coaching voice, a medical monitor beeping, a mayday under a radio check, noise over a drug name, a child's voice placing a bet, a frightened whisper), and with it the correct typed tool call. A words-only null test credits a system only if hearing the call moves its actions more than it moves a pipeline that only reads the words. Only 11 of the 23 systems that can also be run on the transcript pass. Descriptively, errors run toward the words: when the audio calls for protection, all 28 systems carry out the routine request more often than they over-react on clean calls (41% against 12% pooled; the words-only pipeline, 58% against 15%). Exploratory analyses place most of the leading systems' misses on cues they heard; systems beat the null almost entirely on items that state the rule; the leading systems overrule heard resignation or confusion far more often than acute alarm; and, in the models tested, describing the voice and stating the rule each recover part of the shortfall, leaving a gap on emotion.
Comments: 38 pages, 11 figures, 15 tables. Code, scorer and development-split data at this https URL and this https URL
Subjects: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Sound (cs.SD)
ACM classes: I.2.7
Cite as: arXiv:2609.35922 [cs.CL]
  (or arXiv:2609.35922v1 [cs.CL] for this version)
  https://doi.org/10.48550/arXiv.2609.35922

arXiv-issued DOI via DataCite (pending registration)

Submission history

From: Bhavik Mangla [view email]
[v1] Mon, 28 Sep 2026 11:26:06 UTC (522 KB)

来源:HuggingFace Daily Papers(社区热门论文) · arxiv.org