| 电话信道下应用DMFCC进行说话人识别 |
| |
| 引用本文： | 王刚,郑方. 电话信道下应用DMFCC进行说话人识别[J]. 清华大学学报(自然科学版), 2009, 0(10) |
| |
| 作者姓名： | 王刚 郑方 |
| |
| 作者单位： | 清华信息科学与技术国家实验室技术创新与开发部语音和语言技术中心;清华大学计算机科学与技术系; |
| |
| 摘 要： | 鉴别性Mel频率倒谱系数(DMFCC)是一种修正的Mel频率倒谱系数(MFCC),其更加强调语音频谱各个子带携带的鉴别性信息,采用自适应的非均匀的滤波器组设置。在宽带信号应用中,DMFCC的作用和效果已经被证明;但在窄带信号应用中,DMFCC还鲜见有成功应用的例子。该文在电话信道下对应用DMFCC进行说话人识别研究,在美国国家标准技术研究院(NIST)2006年说话人识别评测Female核心测试集上,以MFCC作为特征参数的系统的等错误率为7.57%,以DMFCC作为特征参数的系统的等错误率为7.25%,而采用基于逻辑自回归的线性融合方法把基于两种不同特征的系统在分数域进行融合后系统的等错误率可达到6.31%,相对于基于MFCC的系统等错误率下降16.6%。实验表明,在电话信道下直接应用DMFCC可小幅度提高性能;理论分析以及实验结果表明:二者存在一定的互补性,即把DMFCC和MFCC融合应用能够大幅度提高电话信道下说话人识别的性能。
|
| 关 键 词： | 电话信道 说话人识别 鉴别性Mel频率倒谱系数（DMFCC） 融合 |
Speaker recognition using DMFCC over telephone channels |
| |
| WANG Gang,,ZHENG Thomas Fang. Speaker recognition using DMFCC over telephone channels[J]. Journal of Tsinghua University(Science and Technology), 2009, 0(10) |
| |
| Authors: | WANG Gang ZHENG Thomas Fang |
| |
| Affiliation: | 1.Center for Speech and Language Technologies;Division of Technical Innovation and Development;Tsinghua National Laboratory for Information Science and Technology;Beijing 100084;China;2.Department of Computer Science and Technology;Tsinghua University;China |
| |
| Abstract: | Discriminative Mel frequency cepstrum coefficients(DMFCCs) are modified Mel frequency cepstrum coefficients(MFCC) which emphasize discriminative information carried by sub-bands of the audio spectrum with adaptive non-uniform filter bank settings.The effects of DMFCC have been proven in wide-band signal applications,but not in narrow-band signal applications.This study analyzes the use of DMFCC for speaker recognition over telephone channels.With the NIST Speaker Recognition Evaluation 2006 Female core test... |
| |
| Keywords: | |
| 本文献已被 CNKI 等数据库收录！ |
