Title
Subband-crosscorrelation analysis for robust speech recognition
Abstract
This paper describes subband-crosscorrelation (SBXCOR) analysis using two channel signals. The SBXCOR analysis is an extended signal processing technique of subband-autocorrelation (SBCOR) analysis that extracts periodicities present in speech signals. In this paper, the performance of SBXCOR is investigated using a DTW word recognizer, under simulated acoustic conditions on computer and a real environmental condition. Under the simulated condition, it is assumed that speech signals in each channel are perfectly syn- chronized while noises are not correlated. Consequently, the ef- fective signal-to-noise ratio of the signal generated by simply sum- ming the two signals is raised about 3dB. In such a case, it is shown that SBXCOR is less robust than SBCOR extracted from the two- channel-summed signal, but more robust than the conventional one- channel SBCOR. The resultant performance was much better than that of smoothed group delay spectrum and mel-frequency cepstral coefficient. In a real computer room, it is shown that SBXCOR is more robust than the two-channel-summed SBCOR.
Year
DOI
Venue
1996
10.1109/ICSLP.1996.607144
Spoken Language, 1996. ICSLP 96. Proceedings., Fourth International Conference
Keywords
Field
DocType
cepstral analysis,correlation methods,noise,signal processing,speech recognition,synchronisation,DTW word recognizer,SBCOR,SBXCOR,mel-frequency cepstral coefficient,noise,performance,periodicities,robust speech recognition,signal processing technique,signal-to-noise ratio,simulated acoustic conditions,smoothed group delay spectrum,speech signals,subband-crosscorrelation analysis,synchronization,two-channel-summing signal
Signal processing,Speech processing,Synchronization,Computer science,Cepstrum,Communication channel,Group delay and phase delay,Speech recognition,Cepstral analysis
Conference
Volume
ISBN
Citations 
1
0-7803-3555-4
0
PageRank 
References 
Authors
0.34
4
3
Name
Order
Citations
PageRank
Shoji Kajita114721.92
Kazuya Takeda21301195.60
Fumitada Itakura343167.73