Paper Info
Title
Signal trajectory based noise compensation for robust speech recognition
Abstract
This paper presents a novel signal trajectory based noise compensation algorithm for robust speech recognition. Its performance is evaluated on the Aurora 2 database. The algorithm consists of two processing stages: 1) noise spectrum is estimated using trajectory auto-segmentation and clustering, so that spectral subtraction can be performed to roughly estimate the clean speech trajectories; 2) these trajectories are regenerated using trajectory HMMs, where the constraint between static and dynamic spectral information is imposed to refine the noise subtracted trajectories both in “level” and “shape”. Experimental results show that the recognition performance after spectral subtraction is improved with or without trajectory regeneration, but the HMM regenerated trajectories yields the best performance improvement. After spectral subtraction, the average relative error rate reductions of clean and multi-condition training are 23.21% and 5.58%, respectively. And the proposed trajectory regeneration algorithm further improves them to 42.59% and 15.80%.
Year
DOI
Venue
2006
10.1007/11939993_37
ISCSLP
Keywords
Field
DocType
trajectory regeneration,noise compensation,novel signal trajectory,best performance improvement,proposed trajectory regeneration algorithm,robust speech recognition,dynamic spectral information,trajectory auto-segmentation,trajectory hmms,spectral subtraction,trajectories yield,clean speech trajectory,spectrum,relative error
Speech processing,Pattern recognition,Computer science,Markov model,Speech recognition,Artificial intelligence,Cluster analysis,Hidden Markov model,Subtraction,Trajectory,Approximation error,Performance improvement
Conference
Volume
ISSN
ISBN
4274
0302-9743
3-540-49665-3
Citations 
PageRank 
References 
0
0.34
3
Authors
4
Name
Order
Citations
PageRank
Zhi-Jie Yan17714.34
Jian-Lai Zhou218420.85
Frank K. Soong31395268.29
Ren-Hua Wang434441.36