Paper Info
Title
Variational noise model composition through model perturbation for robust speech recognition with time-varying background noise
Abstract
This study proposes a novel model composition method to improve speech recognition performance in time-varying background noise conditions. It is suggested that each element of the cepstral coefficients represents the frequency degree of the changing components in the envelope of the log-spectrum. With this motivation, in the proposed method, variational noise models are formulated by selectively applying perturbation factors to the mean parameters of a basis model, resulting in a collection of noise models that more accurately reflect the natural range of spectral patterns seen in the log-spectral domain. The basis noise model is obtained from the silence segments of the input speech. The perturbation factors are designed separately for changes in the energy level and spectral envelope. The proposed variational model composition (VMC) method is employed to generate multiple environmental models for our previously proposed parallel combined gaussian mixture model (PCGMM) based feature compensation algorithm. The mixture sharing technique is integrated to reduce computational expenses, caused by employing the variational models. Experimental results prove that the proposed method is considerably more effective at increasing speech recognition performance in time-varying background noise conditions, with +31.31%, +10.65%, and +20.54% average relative improvements in word error rate for speech babble, background music, and real-life in-vehicle noise conditions respectively, compared to the original basic PCGMM method.
Year
DOI
Venue
2011
10.1016/j.specom.2010.12.001
Speech Communication
Keywords
Field
DocType
basis noise model,robust speech recognition,basis model,feature compensation,real-life in-vehicle noise condition,time-varying background noise condition,noise model,time-varying noise,speech recognition performance,multiple environmental models,gaussian mixture model,perturbation factor,variational noise model,variational noise model composition,variational model composition (vmc),model perturbation,speech recognition,energy levels,word error rate,spectrum
Value noise,Speech processing,Background noise,Spectral envelope,Pattern recognition,Noise measurement,Computer science,Word error rate,Speech recognition,Artificial intelligence,Gaussian noise,Gradient noise
Journal
Volume
Issue
ISSN
53
4
Speech Communication
Citations 
PageRank 
References 
2
0.36
17
Authors
2
Name
Order
Citations
PageRank
Wooil Kim112016.95
John H. L. Hansen23215365.75