| 基于VQ的说话人自动识别系统的实现 |
| |
| 引用本文： | 桂苹,吴镇扬,赵力,王维新.基于VQ的说话人自动识别系统的实现[J].电声技术,2003(10):11-14. |
| |
| 作者姓名： | 桂苹 吴镇扬 赵力 王维新 |
| |
| 作者单位： | 东南大学,无线电工程系,江苏,南京,210096 |
| |
| 基金项目： | 国家自然科学基金项目(60272044). |
| |
| 摘 要： | 文中以语音信号的LPC倒谱系数、△倒谱系数、基音周期和△基音周期的混合特征参数作为识别说话人的特征，运用VQ技术实现了说话人自动识别。在10个说话人，1800个汉语数字和单词语音的语音库上进行了系统的识别实验，其中单音节语音的平均识别率达到了92％，双音节语音达到了96．67％，四音节语音达到了97．67％。
|
| 关 键 词： | 声音识别 矢量量化 LPC倒谱系数 △倒谱系数 混合特征 特征提取 |
| 修稿时间： | 2003年4月10日 |
Realization of an automatic speaker recognition system based on VQ methods |
| |
| GUI Ping,WU Zhen-yang,et al..Realization of an automatic speaker recognition system based on VQ methods[J].Audio Engineering,2003(10):11-14. |
| |
| Authors: | GUI Ping WU Zhen-yang |
| |
| Affiliation: | GUI Ping,WU Zhen-yang,et al. |
| |
| Abstract: | This paper presents the realization of a text-dependent speaker recognition system based on VQ methods by using LPC-derived cepstral coefficients,delta-cepstral coefficients,pitches and delta-pitches as the feature parameter set.Through the test of a speech library composed of10speakers and1800speeches of Chi nese numbers and words,fairly high recognition rates,which is about 92%for one-syllable words,96.67%for two-syllable words and97.67%for four-syllable words are reached. |
| |
| Keywords: | speaker recognition vector quantization LPC-derived cepstral coefficient |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
