...值聚类；说话人辨识 [gap=806]Key words: gaussian mixture model (GMM)；F-Ratio; K-means clustering；speaker identification ...
基于24个网页-相关网页
So the main object is to improve the identification rate under noisy environment and the robust performance of speaker identification system.
因此,本文研究的重点就是提高噪声环境下的说话人识别率和说话人辨识系统的鲁棒性能。
参考来源 - 面向短波通信的文本无关说话人辨识研究·2,447,543篇论文数据，部分数据来源于NoteExpress
重点介绍了说话人辨识系统的设计与实现方法。
Emphasis is put on the methods of design and implementation of speaker identification sys - tern.
在文中，音素持续时间信息被添加到传统模型上，以提高说话人辨识率。
In this paper, phoneme duration information is added to the conventional model to improve the recognition rate.
利用听觉频率非线性特性的美尔倒谱作为语音识别的特征参数，来辨识说话人提供的输入口令。
Also, since MFCC represent hearing frequency nonlinear characteristic, we utilize MFCC to be another speak recognition characteristic parameter to distinguish the input passwords.
应用推荐