arXiv 2606.032832026-06-28eess.AScs.SD

SpeakerCard-1M：面向野外说话人确认的基于证据的说话人卡片语料库

SpeakerCard-1M: An Evidence-Grounded Corpus for In-the-Wild Speaker Verification

Brno University of Technology, Czechia（布拉格技术大学，捷克）；Peking University, China（北京大学，中国）；Xiaomi, China（小米，中国）；Athens University of Economics and Business, Greece（雅典经济与商业大学，希腊）；The Hong Kong Polytechnic University, Hong Kong（香港理工大学，香港）；Nanjing University, China（南京大学，中国）

Junyi Peng, Oldřich Plchot, Xiao Song, Dading Chong, Lichun Fan, Hang Su, Themos Stafylakis, Junjie Li, Kong Aik Lee, Shuai Wang, Jian Luan, Jan Černocký

备注：: Corpus and protocols at https://junyipeng00.github.io/SpeakerCard-1M-page

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

提出SpeakerCard-1M双语说话人资源，通过声学探针和受限LLM生成结构化说话人卡片，并定义跨模态协议以支持基于证据的说话人确认。

摘要：

现代说话人确认（SV）系统依赖于说话人嵌入，这些嵌入有效但难以解释或通过自然语言查询。大多数现有的语音-文本语料库针对可控合成或话语级字幕，并为野外说话人识别提供有限的说话人级监督。本文介绍了SpeakerCard-1M，一个面向基于证据的SV的双语说话人中心资源，源自VoxCeleb1/2和CN-Celeb1/2，其中“-1M”后缀指发布中包含的178万条话语级字幕。我们采用工具优先、LLM最后的策略：十个声学探针产生字段级证据，证据在将相对稳定特征与话语级状态分离的模式下聚合成说话人档案，并由仅看到结构化字段的受限LLM渲染出双语说话人卡片。发布内容包括10200个说话人的56700条说话人卡片记录、178万条话语级字幕以及说话人ID不相交的难负三元组。我们进一步定义了两个面向SV的跨模态协议：双向说话人-文本检索（T2S-R / S2T-R）和属性条件验证（AC-Verify），并在零样本强制选择设置下将双编码器基线与最近的音频语言模型进行比较。联合音频-文本训练使VoxCeleb1-O的EER比纯音频基线绝对提高0.31%。在风格对称的LLM生成反事实协议下，八个最近的音频语言模型（7B-30B+参数，包括开源和闭源）在双向强制选择的音高级AC-Verify中得分为49-77%，而我们的双编码器达到88.66%。

英文摘要：

Modern speaker verification (SV) systems rely on speaker embeddings that are effective but difficult to interpret or query in natural language. Most existing speech-text corpora target controllable synthesis or utterance-level captioning, offering limited speaker-level supervision for in-the-wild speaker recognition. This paper introduces SpeakerCard-1M, a bilingual speaker resource for evidence-grounded SV, derived from VoxCeleb1/2 and CN-Celeb1/2, where the ``-1M'' suffix refers to the 1.78M utterance-level captions contained in the release. We adopt a tool-first, LLM-last approach in which ten acoustic probes produce field-level evidence, the evidence is aggregated into speaker profiles under a schema that separates relatively stable traits from utterance-level states, and bilingual Speaker Cards are rendered by a constrained LLM that sees only the structured fields. The release includes 56.7k Speaker Card records over 10.2k speakers, 1.78M utterance-level captions, and speaker-ID-disjoint hard-negative triplets. We further define two SV-oriented cross-modal protocols, bidirectional Speaker-Text Retrieval (T2S-R / S2T-R) and Attribute-Conditioned Verification (AC-Verify), and compare a dual-encoder baseline against recent audio language models under a zero-shot forced-choice setting. Joint audio-text training costs only 0.31% absolute EER on VoxCeleb1-O relative to the audio-only baseline. Under a style-symmetric LLM-generated counterfactual protocol, eight recent audio language models (7B-30B+ parameters, both open- and closed-source) score 49-77% on pitch-level AC-Verify in a 2-way forced-choice setting, compared with 88.66% for our dual encoder.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。