Abstract | ||
---|---|---|
There are two problems when conditional T-S fuzzy neural network is used directly in speech recognition system. One is the rule disaster problem, that is, the rule number will increase exponentially with the increase of input dimensions. Another problem is the network reasoning failure resulted from input dimensions too large. The paper presented an improved algorithm of T-S fuzzy neural network. The subtraction clustering algorithm was used to make certain rule number to escape the rule disaster. The network reasoning can correctly work by adding a compensated factor on membership. The improved algorithm was used in speech recognition system. The experimental results showed that the recognition results of improved algorithm are better than the ones of radial basis function (RBF) neural network using K-means clustering algorithm to select the centroid. And it has much better robustness. |
Year | DOI | Venue |
---|---|---|
2008 | 10.1109/ICNC.2008.404 | ICNC |
Keywords | Field | DocType |
rule disaster problem,input dimension,certain rule number,speech recognition system,neural network,network reasoning failure,fuzzy neural network applied,rule disaster,improved algorithm,network reasoning,t-s fuzzy neural network,k means clustering algorithm,artificial neural networks,clustering algorithms,k means clustering,fuzzy neural network,indexes,cognition,speech recognition,radial basis function | Radial basis function,Computer science,Robustness (computer science),Time delay neural network,Artificial intelligence,Artificial neural network,Cluster analysis,k-means clustering,Pattern recognition,Speech recognition,Centroid,Machine learning,Neural gas | Conference |
Citations | PageRank | References |
0 | 0.34 | 2 |
Authors | ||
4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Xueying Zhang | 1 | 2 | 2.08 |
Peng Wang | 2 | 0 | 0.34 |
Gaoyun Li | 3 | 1 | 1.06 |
Wenjun Hou | 4 | 0 | 9.46 |