...
首页> 外文期刊>Advances in electronic and electric engineering >Performance Analysis of Hindi Voice for Speaker Recognition and Verification Using Different Feature Extraction
【24h】

Performance Analysis of Hindi Voice for Speaker Recognition and Verification Using Different Feature Extraction

机译:不同特征提取的扬声器识别和验证的印地语谱性能分析

获取原文
获取原文并翻译 | 示例
           

摘要

Hindi is the most common language in India and being spoken by about 80% and understanding rate is about 93% by Indians. Different region has many common languages and they pronounce according to their culture environment. Pronunciations may be different but the meaning of the word being spoken is same. So, we try through this paper with the help of different Hindi voice database taken from different part of India and focus to analyze the performance of voice using different feature extraction techniques. This will to identify the proper feature extraction method according to their voices and analyze the performance of these feature extraction techniques. The efficiency of these feature extraction techniques will also help to analyze recognize and verification of the speaker performance.
机译:印地语是印度最常见的语言,大约80%的人讲话,理解率约为印度人约为93%。 不同的地区有许多常用语言,并根据他们的文化环境发音。 发音可能是不同的,但是被说话的词的含义是相同的。 因此,我们在不同的印地文语音数据库中获取本文,从印度的不同部分采取,并专注于使用不同的特征提取技术分析声音的性能。 这将根据其声音识别适当的特征提取方法,并分析这些特征提取技术的性能。 这些特征提取技术的效率也有助于分析识别和验证说话者性能。

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