| 基于MFCC等组合特征的说话人识别模型 |
| |
| 引用本文： | 朱建伟,孙水发,刘晓丽.基于MFCC等组合特征的说话人识别模型[J].三峡大学学报(自然科学版),2009,31(6):77-79,93. |
| |
| 作者姓名： | 朱建伟 孙水发 刘晓丽 |
| |
| 作者单位： | 1. 三峡大学,电气信息学院,湖北,宜昌,443002
2. 三峡大学,电气信息学院,湖北,宜昌,443002;三峡大学,智能视觉与图像信息研究所,湖北,宜昌,443002 |
| |
| 基金项目： | 湖北省教育厅科学技术研究计划重大项目 |
| |
| 摘 要： | 为了有效提取语音特征,提高说话人识别的准确率,系统采用基于有限状态机的端点检测算法对原始语音做VAD处理,提出了新的特征组合参数：基于人的听觉特性的MFCC参数、基于发音生理特征的基音轮廓特征以及衍生的基音周期一阶差分、基音周期变化率,并将它们作为说话人识别系统的特征参数,建立了基于VQ的识别模型.实验表明：本文系统使用VAD,使系统的识别率提高了5%8%,较单独使用MFCC参数的说话人识别系统的识别率提高了2%3%.
|
| 关 键 词： | 说话人识别 Mel倒谱系数 基音轮廓特征 语音活性检测 |
Speaker Recognition Model Based on MFCC and Combined Features |
| |
| Zhu Jianwei,Sun Shuifa,Liu Xiaoli.Speaker Recognition Model Based on MFCC and Combined Features[J].Journal of China Three Gorges University(Natural Sciences),2009,31(6):77-79,93. |
| |
| Authors: | Zhu Jianwei Sun Shuifa Liu Xiaoli |
| |
| Affiliation: | Zhu Jianwei Sun Shuifa Liu Xiaoli(1. College of Electrical Engineering ＆ Information Science, China Three Gorges Univ. , Yichang 443002, China;2. Institute of Intelligent Vision and Image Information, China Three Gorges Univ. , Yiehang 443002,China) |
| |
| Abstract: | In order to effectively extract the speech features and improve the speaker recognition accuracy,a VAD algorithm based on the finite state machine（FSM） is applied on the original voice firstly.The following four features are selected： the Mel frequency cepstral coefficient（MFCC） parameters based on the characteristics of human hearing,the pitch contour based on the physiological characteristics of pronunciation features,the pitch first-order difference and the pitch changed rate.Vector quantization（VQ） based speaker recognition model is established.The experimental results show that the recognition rate of the proposed system is improved 2%-3% than the speaker recognition system using the MFCC parameters only,and 5%-8% than the system without using the VAD. |
| |
| Keywords: | speaker recognition MFCC pitch contour features VAD |
| 本文献已被 维普 万方数据 等数据库收录！ |
|
