ESD 情感语音数据集

ESD 全称 Emotional Speech Database，是一个用于语音转换研究的情感语音数据集。数据集由 10 个以英语为母语和 10 个以汉语为母语的人说的 350 个平行话语组成，涵盖 5 个情感类别（中性、快乐、愤怒、悲伤和惊讶）。在受控声学环境中记录了超过 29 小时的语音数据。该数据集适用于多语言和跨语言情感语音转换研究。

ESD, the abbreviation of Emotional Speech Database, is an emotional speech dataset designed for speech conversion research. It comprises 350 parallel utterances spoken by 10 native English speakers and 10 native Mandarin speakers, covering five emotional categories: neutral, happy, angry, sad, and surprised. Over 29 hours of speech data were recorded in a controlled acoustic environment. This dataset is applicable to multilingual and cross-lingual emotional speech conversion research.

搜集汇总

背景与挑战

背景概述

ESD情感语音数据集是一个多语言情感语音数据集，包含英语和汉语母语者录制的350个平行话语，涵盖5种情感类别（中性、快乐、愤怒、悲伤和惊讶），总时长超过29小时，适用于情感语音转换研究。