Title | ||
---|---|---|
A speech enhancement method based on sparse reconstruction of power spectral density. |
Abstract | ||
---|---|---|
•The approximation K-singular value decomposition algorithm with nonnegative constraint is used to train PSD dictionary.•The least angle regression algorithm with a new termination rule is applied to obtain the sparse representation.•The termination rule is related to the noise level and nonzero cross terms of the speech and noise spectra.•The enhanced speech signal is obtained by using the estimated PSD and subspace approach. |
Year | DOI | Venue |
---|---|---|
2014 | 10.1016/j.compeleceng.2013.12.007 | Computers & Electrical Engineering |
Field | DocType | Volume |
Speech enhancement,Rule-based system,Pattern recognition,Subspace topology,Computer science,Sparse approximation,Spectral density,Artificial intelligence,Norm (mathematics),Least-angle regression,Signal subspace | Journal | 40 |
Issue | ISSN | Citations |
4 | 0045-7906 | 1 |
PageRank | References | Authors |
0.36 | 18 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Yanping Zhao | 1 | 1 | 1.37 |
Xiaohui Zhao | 2 | 87 | 15.89 |
Bo Wang | 3 | 67 | 8.19 |