KSUEmotions

<h3>Introduction</h3><br> <p>KSUEmotions was developed by <a href="http://ksu.edu.sa/en/">King Saud University</a> (KSU) and contains approximately five hours of emotional Modern Standard Arabic (MSA) speech from 23 subjects. Speakers were from three countries: Yemen, Saudi Arabia and Syria.</p><br> <p>Subjects read MSA sentences from newswire text in the following emotions: neutral, anger, sadness, happiness, surprise, and interrogative (asking a question). Human reviewers then listened to the recordings to identify the emotion they heard..</p><br> <h3>Data</h3><br> <p>Audio was recorded in each participant's home. Audio is presented as 16-bit 16 kHz flac compressed wav. In addition to speech files and metadata about the speakers, timeless label files and automatic time segmentation alignment files are included. Text is presented as UTF-8 plain text.</p><br> <h3>Samples</h3><br> <p>Please listen to this <a href="desc/addenda/LDC2017S12.flac">neutral audio sample</a>.</p><br> <h3>Updates</h3><br> <p>None at this time.</p></br> Portions © 2017 Ali Meftah, © 2017 Trustees of the University of Pennsylvania<h3>数据集简介</h3><br> <p>KSUEmotions 由<a href="http://ksu.edu.sa/en/">沙特国王大学（King Saud University，简称KSU）</a>研发，包含来自23名受试者的约5小时情感类现代标准阿拉伯语（Modern Standard Arabic，简称MSA）语音数据。受试者均来自也门、沙特阿拉伯与叙利亚三个国家。</p><br> <p>受试者需朗读源自新闻专线文本的现代标准阿拉伯语语句，涵盖以下情感类别：中性、愤怒、悲伤、喜悦、惊讶，以及疑问（提问式）。随后由人工评审员收听录音，标注其感知到的情感。</p><br> <h3>数据说明</h3><br> <p>音频均在每位参与者的家中录制，格式为16位、16 kHz的FLAC压缩波形音频。除语音文件与发音者元数据外，数据集还包含时序标注文件与自动时间分段对齐文件。文本内容采用UTF-8纯文本格式存储。</p><br> <h3>样本示例</h3><br> <p>请收听此<a href="desc/addenda/LDC2017S12.flac">中性语音样本</a>。</p><br> <h3>更新记录</h3><br> <p>暂无更新记录。</p></br> 部分内容 © 2017 Ali Meftah，© 2017 宾夕法尼亚大学托管委员会

搜集汇总

背景与挑战

背景概述

KSUEmotions是一个现代标准阿拉伯语情感语音数据集，包含约5小时来自23名说话者的录音，覆盖六种基本情感（中性、愤怒、悲伤、快乐、惊讶、疑问），数据经过人工评审验证，适用于语音情感识别研究。