Abstract | ||
---|---|---|
A new algorithm based on information maximization is proposed, against the shortcomings of traditional Speech Blind separation of low convergence rate and high crosstalk error. It uses the new sign function to make mutual information of input and output to maximize by analyzing a variety of non-linear function of the separation performance. Experiments show that the new algorithm has advantages, such as fast convergence, small crosstalk error and good separation efficiency which compared with traditional methods of Be11's and Sejnowskl's. |
Year | DOI | Venue |
---|---|---|
2010 | 10.1109/ICIE.2010.25 | Journal of Networks |
Keywords | DocType | Volume |
information maximum,crosstalk error,speech processing,high crosstalk error,modified speech blind separation,information,low convergence rate,maximum,blind separation,new sign function,blind source separation,information maximization,separation performance,good separation efficiency,mutual information,new algorithm,traditional speech blind separation,speech,modified speech blind separation method,fast convergence,convergence rate,transfer functions,crosstalk,algorithm design and analysis | Journal | 1 |
Issue | ISBN | Citations |
12 | 978-1-4244-7507-0 | 1 |
PageRank | References | Authors |
0.36 | 2 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Xueying Zhang | 1 | 38 | 9.52 |
Hairong Jia | 2 | 1 | 0.70 |
Hong Xu | 3 | 1 | 0.36 |