Affiliation:
1. Guilin University of Electronic Technology
Abstract
Due to MFCC characteristic parameter in speech recognition has low identification accuracy when signal is intermediate, high frequency signal, this paper put forward a improved algorithm of combining MFCC, Mid-MFCC and IMFCC, using increase or decrease component method to calculate the contribution that MFCC, Mid-MFCC and IMFCC each order cepstrum component was used in speech emotion recognition, extracting several order cepstrum component with highest contribution from three characteristic parameters and forming a new characteristic parameter. The experiment results show that under the same environment new characteristic parameter has higher recognition rate than classic MFCC characteristic parameter in speech emotion recognition.
Publisher
Trans Tech Publications, Ltd.
Reference6 articles.
1. Zhen Bin, Wu XiHong, Liu ZhiMin, Chi HuiSheng. On the Importance of Components of the MFCC in Speech and Speaker Recognition[J]. Acta Scientiarum Naturalium Universitatis Pekinensis, 2001, 37(3): 371-378.
2. Lv XiaoYun, Wang HongXia. Abnormal audio recognition algorithm based on MFCC and short-term energy[J]. Journal of Computer Applications, 2010, 30(3): 796-798.
3. Han Yi, Wang GuoYin, Yang Yong. Speech emotion recognition based on MFCC[J]. Journal of Chongqing University of Posts and Telecommunications (Natural Science), 2008, 20(5): 507-602.
4. Mao Xia, Chen LiJiang. Extraction and Modeling Method of Speech Emotion Informatio[P]. China patent: CN101261832, 2008-4-21.
5. Mao Xia, Chen LiJiang. Speech Emotion Recognition Based on Parametric Filter and Fractal Dimension1[J]. IEICE TRANSACTIONS ON INFORMATION AND SYSTEMS (SCIE Index, IF: 0. 396). 2010, 93(8): 2324-2326.
Cited by
6 articles.
订阅此论文施引文献
订阅此论文施引文献，注册后可以免费订阅5篇论文的施引文献，订阅后可以查看论文全部施引文献