...
  • 主题
高级搜索  >
历史搜索 清空历史
首页> 外文期刊>The Journal of grey system >Text-Independent speaker recognition based on one third octave feature and grey relational analysis
【24h】

Text-Independent speaker recognition based on one third octave feature and grey relational analysis

机译：基于三分之一八度音阶特征和灰色关联分析的与文本无关的说话人识别

获取原文
获取原文并翻译 | 示例
           
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Aiming at the problem that, among the traditional speaker recognition methods, the data of speech signal feature parameters is large and its modeling takes longer time, a text-independent speaker recognition method, based on one third octave feature parameters of speakers' speech signals and grey relational analysis, is put forward. The comparative data sequences are established based on one third octave feature parameters of the speech signals of the registered speakers and the to-be identified speakers, then use the one third octave feature parameters of the to-be identified speaker's speech signal as the reference data sequence, and then extract the maximum grey correlation degree and compare it with the set correlation degree threshold, finally realize the speaker recognition by the comparative results. Recognition experimental results show that the method proposed in this paper is simple and effective, achieving higher accuracy of speaker recognition.
机译：针对传统说话人识别方法中语音信号特征参数的数据量大，建模时间长的问题，提出了一种基于说话人语音信号三分之一八度特征参数的文本无关说话人识别方法。提出了灰色关联分析。基于注册讲话者和待识别讲话者的语音信号的三分之一八度特征参数建立比较数据序列，然后将待识别讲话者的语音信号的三分之一八度特征参数作为参考数据序列，然后提取最大的灰色关联度并将其与设置的关联度阈值进行比较，最后通过比较结果实现说话人识别。识别实验结果表明，本文提出的方法简单有效，实现了说话人识别的较高准确性。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号