TY - GEN
T1 - Low-order volterra long-term predictors
AU - Despotovic, Vladimir
AU - Görtz, Norbert
AU - Peric, Zoran
N1 - Publisher Copyright:
© VDE VERLAG GMBH • Berlin • Offenbach.
PY - 2012
Y1 - 2012
N2 - Models based on linear prediction have been used for several decades in different areas of speech signal processing. While the linear approach has led to great advances in the last 40 years, it neglects nonlinearities present in the speech production mechanism. This paper compares the results of long-term nonlinear prediction based on second-order and third-order Volterra filters. Additional improvement can be obtained using fractional-delay long-term prediction. Experimental results reveal that the proposed method outperforms linear long-term prediction techniques in terms of prediction gain.
AB - Models based on linear prediction have been used for several decades in different areas of speech signal processing. While the linear approach has led to great advances in the last 40 years, it neglects nonlinearities present in the speech production mechanism. This paper compares the results of long-term nonlinear prediction based on second-order and third-order Volterra filters. Additional improvement can be obtained using fractional-delay long-term prediction. Experimental results reveal that the proposed method outperforms linear long-term prediction techniques in terms of prediction gain.
UR - http://www.scopus.com/inward/record.url?scp=84947925870&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:84947925870
T3 - Proceedings of 10th ITG Symposium on Speech Communication
SP - 27
EP - 30
BT - Proceedings of 10th ITG Symposium on Speech Communication
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 10th ITG Symposium on Speech Communication, ITGspeech 2012
Y2 - 26 September 2012 through 28 September 2012
ER -