Speech Smoothing Algorithm Based On MFCC

摘要：

Because the naturalness of Chinese speech synthesis is not perfect, this paper put forward an optimal algorithm, which is Chinese speech spectral smoothing algorithm based on Mel-scale Frequency Cepstral Coefficients(MFCC). This smoothing algorithm not only smoothes the speech spectrum,but also avoids degrading of the natural degree. The experiments show that Chinese speech synthesis has higer quality which is caused by this algorithm.

关键词： Speech synthesis syste Spectral Smoothing MFCC

作者: Jiali Wang Xueyuan Lin Yazhou Zhang Famai Liang

作者单位: Department of Electronic Engineering, Naval Aeronautical and Astronautical University,Yantai, China

会议类型: 国际会议

会议名称: The Third International Conference on Modelling and Simulation(第三届国际建模、计算、仿真、优化及其应用学术会议 ICMS 2010)

会议地点: 无锡

会议语种:英文

页码: 286-289

在线出版日期: 2010-06-04（万方平台首次上网日期，不代表论文的发表时间）