Improved MFCC-based feature for robust speaker identification | |
Wu Zunjing ; Cao Zhigang | |
2010-05-06 ; 2010-05-06 | |
关键词 | Theoretical or Mathematical/ cepstral analysis feature extraction median filters speaker recognition speech enhancement/ MFCC-based feature robust speaker identification mel-frequency cepstral coefficient speech feature noise interference recognition systems logarithmic transformation MFCC analysis noise sensitivity feature extraction speech enhancement spectral subtraction median-filter noise suppression recognition error rate signal-to-noise ratio/ B6130E Speech recognition and synthesis C5260S Speech processing techniques |
中文摘要 | The mel-frequency cepstral coefficient (MFCC) is the most widely used feature in speech and speaker recognition. However, MFCC is very sensitive to noise interference, which tends to drastically degrade the performance of recognition systems because of the mismatches between training and testing. In this paper, the logarithmic transformation in the standard MFCC analysis is replaced by a combined function to improve the noisy sensitivity. The proposed feature extraction process is also combined with speech enhancement methods, such as spectral subtraction and median-filter to further suppress the noise. Experiments show that the proposed robust MFCC-based feature significantly reduces the recognition error rate over a wide signal-to-noise ratio range. |
语种 | 英语 ; 英语 |
出版者 | Editorial Board of J. of Tsinghua University ; China |
内容类型 | 期刊论文 |
源URL | [http://hdl.handle.net/123456789/12070] |
专题 | 清华大学 |
推荐引用方式 GB/T 7714 | Wu Zunjing,Cao Zhigang. Improved MFCC-based feature for robust speaker identification[J],2010, 2010. |
APA | Wu Zunjing,&Cao Zhigang.(2010).Improved MFCC-based feature for robust speaker identification.. |
MLA | Wu Zunjing,et al."Improved MFCC-based feature for robust speaker identification".(2010). |
个性服务 |
查看访问统计 |
相关权益政策 |
暂无数据 |
收藏/分享 |
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。
修改评论