作为非典型语音和情感语音风格可解释基元的语音质量维度
Voice Quality Dimensions as Interpretable Primitives for Speaking Style for Atypical Speech and Affect
AI总结：
本文针对七个语音和话语维度开发并评估了语音质量模型，在SAP数据集上训练探针，验证了其出色的跨类别泛化能力，并在多语言非典型语音和情感语音数据集上实现了出色的零样本性能，证明了语音质量维度在语音风格任务中的实用性。
摘要：
感知语音质量维度描述了非典型语音及其他语音调制的关键特征。本文针对七个语音和话语维度（清晰度、辅音不精准、粗糙声、自然度、单一响度、单一音调和气息声）开发并评估了语音质量模型。探针在公开的 Speech Accessibility (SAP) 项目数据集上进行训练，该数据集包含来自434名说话人的11,184个样本，并使用冻结的预训练模型的嵌入作为特征。我们发现，我们的探针在SAP数据集的各种语音诱发类别中均具有出色的性能和强大的泛化能力。我们进一步在额外数据集上验证了零样本性能，涵盖未见过的语言和任务：意大利语非典型语音、英语非典型语音和情感语音。出色的零样本性能以及在一系列评估中结果的可解释性，表明了在语音风格相关任务中使用语音质量维度的实用性。
英文摘要：
Perceptual voice quality dimensions describe key characteristics of atypical speech and other speech modulations. Here we develop and evaluate voice quality models for seven voice and speech dimensions (intelligibility, imprecise consonants, harsh voice, naturalness, monoloudness, monopitch, and breathiness). Probes were trained on the public Speech Accessibility (SAP) project dataset with 11,184 samples from 434 speakers, using embeddings from frozen pre-trained models as features. We found that our probes had both strong performance and strong generalization across speech elicitation categories in the SAP dataset. We further validated zero-shot performance on additional datasets, encompassing unseen languages and tasks: Italian atypical speech, English atypical speech, and affective speech. The strong zero-shot performance and the interpretability of results across an array of evaluations suggests the utility of using voice quality dimensions in speaking style-related tasks.
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。
