TY - GEN
T1 - Blind determination of the signal to noise ratio of speech signals based on estimation combination of multiple features
AU - Ondusko, Russell
AU - Marbach, Matthew
AU - McClellan, Andrew
AU - Ramachandran, Ravi P.
AU - Head, Linda M.
AU - Huggins, Mark C.
AU - Smolenski, Brett Y.
PY - 2006
Y1 - 2006
N2 - A blind approach for estimating the signal to noise ratio (SNR) of a speech signal corrupted by additive noise is proposed. The method is based on a pattern recognition paradigm using various linear predictive based features, a vector quantizer classifier and estimation combination. Blind SNR estimation is very useful in speaker identification systems in which a confidence metric is determined along with the speaker identity. The confidence metric is partially based on the mismatch between the training and testing conditions of the speaker identification system and SNR estimation is very important in evaluating the degree of this mismatch. The aim is to correctly estimate SNR values from 0 to 30 dB, a range that is both practical and crucial for speaker identification systems. Additive white Gaussian noise and pink noise are investigated. The best feature for both white and pink noise is the vector of reflection coefficients which achieves an average SNR estimation error of 1.6 dB and 1.85 dB for white and pink noise respectively. Combining the estimates of 4 features lowers the error for white noise to 1.46 dB and for pink noise to 1.69 dB.
AB - A blind approach for estimating the signal to noise ratio (SNR) of a speech signal corrupted by additive noise is proposed. The method is based on a pattern recognition paradigm using various linear predictive based features, a vector quantizer classifier and estimation combination. Blind SNR estimation is very useful in speaker identification systems in which a confidence metric is determined along with the speaker identity. The confidence metric is partially based on the mismatch between the training and testing conditions of the speaker identification system and SNR estimation is very important in evaluating the degree of this mismatch. The aim is to correctly estimate SNR values from 0 to 30 dB, a range that is both practical and crucial for speaker identification systems. Additive white Gaussian noise and pink noise are investigated. The best feature for both white and pink noise is the vector of reflection coefficients which achieves an average SNR estimation error of 1.6 dB and 1.85 dB for white and pink noise respectively. Combining the estimates of 4 features lowers the error for white noise to 1.46 dB and for pink noise to 1.69 dB.
UR - http://www.scopus.com/inward/record.url?scp=50249152884&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=50249152884&partnerID=8YFLogxK
U2 - 10.1109/APCCAS.2006.342229
DO - 10.1109/APCCAS.2006.342229
M3 - Conference contribution
AN - SCOPUS:50249152884
SN - 1424403871
SN - 9781424403875
T3 - IEEE Asia-Pacific Conference on Circuits and Systems, Proceedings, APCCAS
SP - 1895
EP - 1898
BT - APCCAS 2006 - 2006 IEEE Asia Pacific Conference on Circuits and Systems
T2 - APCCAS 2006 - 2006 IEEE Asia Pacific Conference on Circuits and Systems
Y2 - 4 December 2006 through 6 December 2006
ER -