现代电子技术Issue(10)：7-9,15,4.
基于改进C0复杂度和MFCC相似度的端点检测
Endpoint detection based improvement of C0 complexity and MFCC similarity
许昊 1张二华1
作者信息
- 1. 南京理工大学 计算机科学与工程学院，江苏 南京 210094
- 折叠
摘要
Abstract
In order to improve the accuracy of voice activity detection(VAD),an algorithm based on improvement of C0 complexity and Mel frequency cepstrum coefficient(MFCC)similarity is proposed. First,the C0 complexity and MFCC similarity of speech signal in each frame is calculated first,and then the comprehensive judgment is made in combination with the C0 com⁃plexity and MFCC similarity. VAD for noisy speech whose SNR ranges from ⁃15dB to 15dB is conducted. Matlab software is used for simulation experiment. The results show that the accuracy and stability of the proposed method is higher than only C0 complexity detecting method or MFCC similarity detecting method.关键词
音信号处理/C0复杂度/MFCC相似度/端点检测Key words
speech signal processing/C0 complexity/Mel frequency cepstrum coefficient(MFCC)similarity/endpoint de-tection分类
信息技术与安全科学引用本文复制引用
许昊,张二华..基于改进C0复杂度和MFCC相似度的端点检测[J].现代电子技术,2015,(10):7-9,15,4.
