计算机工程与应用 ›› 2008, Vol. 44 ›› Issue (18): 15-16.

• 博士论坛 • 上一篇    下一篇

基于SOM网络的语音训练可视化方法研究

杨 丹1,徐 彬2,王 旭1,廖富成1   

  1. 1.东北大学 信息科学与工程学院,沈阳 110004
    2.东北大学 计算中心,沈阳 110004
  • 收稿日期:2008-02-28 修回日期:2008-03-24 出版日期:2008-06-21 发布日期:2008-06-21
  • 通讯作者: 杨 丹

Research of speech training visual method using Self-Organizing Map

YANG Dan1,XU Bin2,WANG Xu1,LIAO Fu-cheng1   

  1. 1.School of Information Science & Engineering,Northeastern University,Shenyang 110004,China
    2.Computer Center,Northeastern University,Shenyang 110004,China
  • Received:2008-02-28 Revised:2008-03-24 Online:2008-06-21 Published:2008-06-21
  • Contact: YANG Dan

摘要: 提出了一种利用SOM网络输出层可视化的特点进行语音训练的方法。SOM网络能够将输入向量映射到二维平面或曲面上,受试者通过视觉反馈的位置信息,指导其发音行为。为了提高SOM聚类效果,SOM还进行加强训练;讨论了SOM输出层神经元个数对聚类的影响。实验结果表明,提出的利用SOM语音训练方法,直观简单,能够有效地实现“看图说话”。

Abstract: A speech training method using output-layer visualization of Self-Organizing Map(SOM) is proposed.SOM is a neural
network model that can transform input data onto two-dimensional plane or curve surface of output layer neurons.The subjects guide their pronunciation through visual feedback from positional information of output layer neurons.In order to improve the clustering of SOM,the authors make strengthen training and discuss how to choose the number of neurons in the output layer.The results show the proposed speech training method is simple and straightforward.It effectively realizes “speak when seeing the picture”.