DesignLex
/人机交互/Vocal prosody
论文专业

Vocal prosody

语音韵律(Vocal Prosody),指语音中的节奏、重音、语调、音高变化等承载情感与态度等副语言信息的声学特征,是虚拟代理表达情绪的重要模态。

English Definition

"The rhythm, stress, intonation, and melodic patterns of speech that carry paralinguistic information such as emotion and attitude, used as a modality for emotional expression in virtual agents."

用法说明 · 针对中文母语者

在多模态情感表达研究中,vocal prosody 通常与 facial expression 配对构成「面部–语音」双模态条件;中文文献中也可称「声学韵律」。

常见错误

  • ·不要把 prosody 简单译成「韵律」就完事——在 agent 设计语境下要强调其情绪表达功能

真实用例 · 2

"another in which they convey emotions through both facial expressions and vocal prosody."

论文Pipeline extracted — review needed·Introduction 中描述第二个研究条件(多模态反馈)

"outperformed other baseline models under both facial-only and facial-voice conditions"

论文Pipeline extracted — review needed·Abstract 中作为评估条件的对照描述
由 pipeline 自动采集,待人工 review