| 二次特征提取及其在说话人识别中的应用 |
| |
| 引用本文： | 马志友,杨莹春,吴朝晖. 二次特征提取及其在说话人识别中的应用[J]. 电路与系统学报, 2003, 8(2): 130-133 |
| |
| 作者姓名： | 马志友 杨莹春 吴朝晖 |
| |
| 作者单位： | 浙江大学,玉泉校区计算机系统工程研究所,浙江,杭州,310027 |
| |
| 基金项目： | 国家自然科学基金资助项目(60273059)，国家863计划基金资助项目(2001AA4180)，浙江省教育厅基金资助项目(20020721)，浙江省自科基金青年科技人才培养专项基金资助项目(RC01058)，博士点基金资助项目(20020335025) |
| |
| 摘 要： | 传统的特征提取方法在处理小范围的说话人识别时尚可为之，但是在较大用户群的情况下，由于特征覆盖范围不够导致性能下降。鉴于此，本文提出了一种新的二次特征提取方法，它通过综合运用加权、微分、组合、筛选等方法，进一步挖掘说话人语音背后的隐性个性差异。在采用138人的YOHO数据库上进行的说话人识别测试中，其性能优于传统的特征提取方法。
|
| 关 键 词： | 发音机理 说话人识别 特征提取 二次特征提取 MFCC LPCC |
| 文章编号： | 1007-0249(2003)02-0130-04 |
| 修稿时间： | 2002-09-16 |
Further Feature Extraction and Its Application on Speaker Recognition |
| |
| MA Zhi-you,YANG Yin-chun,WU Zhao-hui. Further Feature Extraction and Its Application on Speaker Recognition[J]. Journal of Circuits and Systems, 2003, 8(2): 130-133 |
| |
| Authors: | MA Zhi-you YANG Yin-chun WU Zhao-hui |
| |
| Abstract: | The traditional feature extraction seems good to recognize speaker by their voice in small case. Nevertheless, the performance of recognizer is usually depressed by the limited feature space unable to cover the increasing number of speakers to be recognized. Accordingly, a novel feature extraction method is proposed, in which effective measures such as weight, differential, combination and selection, are taken to explore those voice characteristics that can be used to distinguish different speakers. Experiment based on 138-person YOHO database demonstrates that better performance can be achieved by the proposed method compared with traditional feature extraction methods. |
| |
| Keywords: | Mechanism of Speech Production Speaker Recognition Feature Extraction Further Feature Extraction MFCC LPCC |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
