欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
语音的特征提取是说话人识别系统中的关键问题.在传统的Mel频率倒谱系数（MFCC）参数的基础上,提出一种改进的MFCC特征提取算法.该算法着眼于语音的前端处理,在预处理阶段,利用SWCE窗函数,对信号进行多窗频谱估计.并对得到的频谱进行平滑处理,得到信号的谱包络.然后对信号的谱包络进行计算,得到改进的MFCC参数.实验表明,在不同噪声环境下,与传统的MFCC算法相比,改进的算法识别率提高四个百分点以上.
英文摘要：
Feature extraction of speech is a key problem in speaker recognition system.An improved MFCC feature extraction algorithm which is based on the traditional MFCC parameters was proposed,which focuses on the frontend processing.In the preprocessing stage,the SWCE window function was used to get the multiple window spectrum estimation.And the obtained spectrum was smoothed to get envelope signal.Then the spectrum envelope signal was calculated to get the improved MFCC parameters.Experiments show that in the noisy environment the recognition rate of the new algorithm,compared with the traditional MFCC algorithm,is improved more than four precent.
同期刊论文项目
同项目期刊论文
Parameter optimization and application of support vector machine based on parallel artificial fish s
Study for classification of emotional speech by using optimized frame zero crossing with peak amplit
A robust feature extraction approach based on an auditory model for classification of speech and exp
期刊信息
位置：