In this paper,based on Kohonen's SOFM model, a new learning algorithm is given, in which the lateral feedback effects are strengthened. Experements have shown that this algorithm speeds up the processing of producing a topographically ordered feature map and improves the clustering effect. Keywords: Neural Networks; Self-Organization; Lateral Feedback; Differential Competitive Learning; Phoneme Map
Cite as: Jianxin, J., Kechu, Y., Zheng, H. (1991) A new self-organization algorithm of forming a phoneme map. Proc. 2nd European Conference on Speech Communication and Technology (Eurospeech 1991), 125-128, doi: 10.21437/Eurospeech.1991-27
@inproceedings{jianxin91_eurospeech,
author={Jiang Jianxin and Yi Kechu and Hu Zheng},
title={{A new self-organization algorithm of forming a phoneme map}},
year=1991,
booktitle={Proc. 2nd European Conference on Speech Communication and Technology (Eurospeech 1991)},
pages={125--128},
doi={10.21437/Eurospeech.1991-27}
}