Noisy Speech Recognition Based on Integration/Selection of Multiple Noise Suppression Methods Using Noise GMMs

Norihide KITAOKA  Souta HAMAGUCHI  Seiichi NAKAGAWA  

Publication
IEICE TRANSACTIONS on Information and Systems   Vol.E91-D   No.3   pp.411-421
Publication Date: 2008/03/01
Online ISSN: 1745-1361
DOI: 10.1093/ietisy/e91-d.3.411
Print ISSN: 0916-8532
Type of Manuscript: Special Section PAPER (Special Section on Robust Speech Processing in Realistic Environments)
Category: Noisy Speech Recognition
Keyword: 
noisy speech recognition,  noise suppression method selection,  CENSREC-1,  

Full Text: PDF(879.8KB)>>
Buy this Article




Summary: 
To achieve high recognition performance for a wide variety of noise and for a wide range of signal-to-noise ratio, this paper presents methods for integration of four noise reduction algorithms: spectral subtraction with smoothing of time direction, temporal domain SVD-based speech enhancement, GMM-based speech estimation and KLT-based comb-filtering. In this paper, we proposed two types of combination methods of noise suppression algorithms: selection of front-end processor and combination of results from multiple recognition processes. Recognition results on the CENSREC-1 task showed the effectiveness of our proposed methods.