计算机工程与应用 ›› 2008, Vol. 44 ›› Issue (21): 33-36.DOI: 10.3778/j.issn.1002-8331.2008.21.009
• 博士论坛 • 上一篇 下一篇
黄 挺,杨莹春
收稿日期:
修回日期:
出版日期:
发布日期:
通讯作者:
HUANG Ting,YANG Ying-chun
Received:
Revised:
Online:
Published:
Contact:
摘要： 提出了一种基于情感语音的差异检测与剔除的说话人识别方法，克服了前人的方法中需要在训练时提供测试说话人的情感语音或者需要在测试时提供测试语音的情感状态信息给系统的使用带来的不便性，并在识别性能上比传统的ASR系统提高4.7%。
关键词: 情感语音, 差异检测, 基频包络, 说话人识别
Abstract: This paper presents a new robust speaker recognition system based on difference detection and pruning of emotional speech，which avoids the inconvenience of needing the speaker emotional voice in training or the emotion state of the speaker in testing in other speaker recognition system.Furthermore，compared with the traditional ASR system，it enhances the system performance under emotion speech by 4.7% in identification rate.
Key words: emotion speech, difference detection, pitch envelope, speaker recognition
黄 挺,杨莹春. 基于情感语音的差异检测与剔除的说话人识别[J]. 计算机工程与应用, 2008, 44(21): 33-36.
HUANG Ting,YANG Ying-chun. Applying difference detection and pruning to emotional speaker recognition[J]. Computer Engineering and Applications, 2008, 44(21): 33-36.
0 / 推荐
导出引用管理器 EndNote|Ris|BibTeX
链接本文: http://cea.ceaj.org/CN/10.3778/j.issn.1002-8331.2008.21.009
http://cea.ceaj.org/CN/Y2008/V44/I21/33
