VoxEmo：基于语音大语言模型的语音情感识别基准测试
VoxEmo: Benchmarking Speech Emotion Recognition with Speech LLMs
AI总结：
VoxEmo是一个基于语音大语言模型的语音情感识别基准测试，通过提供多样化的提示复杂度和分布感知的软标签协议，评估模型在真实世界中的情感识别能力。
摘要：
语音大语言模型（LLMs）在通过生成接口进行语音情感识别（SER）方面展现出巨大潜力。然而，从封闭集分类转向开放文本生成引入了零样本随机性，使评估对提示极为敏感。此外，传统语音LLM基准测试忽略了人类情感的固有模糊性。因此，我们提出了VoxEmo，一个涵盖15种语言35个情感语料库的全面SER基准测试，适用于语音LLM。VoxEmo提供了一个标准化工具包，包含从直接分类到语调推理的多种提示复杂度。为了反映现实世界感知/应用，我们引入了分布感知的软标签协议和提示-集合策略，以模拟标注者分歧。实验表明，尽管零样本语音LLM在硬标签准确性上落后于监督基线，但它们能够与人类主观分布独特地对齐。
英文摘要：
Speech Large Language Models (LLMs) show great promise for speech emotion recognition (SER) via generative interfaces. However, shifting from closed-set classification to open text generation introduces zero-shot stochasticity, making evaluation highly sensitive to prompts. Additionally, conventional speech LLMs benchmarks overlook the inherent ambiguity of human emotion. Hence, we present VoxEmo, a comprehensive SER benchmark encompassing 35 emotion corpora across 15 languages for Speech LLMs. VoxEmo provides a standardized toolkit featuring varying prompt complexities, from direct classification to paralinguistic reasoning. To reflect real-world perception/application, we introduce a distribution-aware soft-label protocol and a prompt-ensemble strategy that emulates annotator disagreement. Experiments reveal that while zero-shot speech LLMs trail supervised baselines in hard-label accuracy, they uniquely align with human subjective distributions.
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。
