E. K. Kokkinis, A. Tsilfidis, E. Georganti, and J. Mourjopoulos, “Joint Noise and Reverberation Suppression for Speech Applications,” in Proc. AES Convention 130, May 2011, pp. 8431:1-10. [Online]. Available: https://aes.org/publications/elibrary-page/?id=15898
Kokkinis EK, Tsilfidis A, Georganti E, Mourjopoulos J. Joint Noise and Reverberation Suppression for Speech Applications. In: AES Convention 130. Audio Engineering Society; 2011.pp. 8431:1-10. Available from: https://aes.org/publications/elibrary-page/?id=15898
@inproceedings{Kokkinis2011_15898,
author = {Kokkinis, Elias K. and Tsilfidis, Alexandros and Georganti, Eleftheria and Mourjopoulos, John},
title = {{Joint Noise and Reverberation Suppression for Speech Applications}},
booktitle = {AES Convention 130},
pages = {8431:1-10},
url = {https://aes.org/publications/elibrary-page/?id=15898},
year = {2011},
month = may,
publisher = {Audio Engineering Society},
address = {London Uk}
}
TY - CPAPER
TI - Joint Noise and Reverberation Suppression for Speech Applications
AU - Kokkinis, Elias K.
AU - Tsilfidis, Alexandros
AU - Georganti, Eleftheria
AU - Mourjopoulos, John
T2 - AES Convention 130
SP - 1
EP - 10
M1 - Paper 8431
PY - 2011
DA - 2011/05/06
UR - https://aes.org/publications/elibrary-page/?id=15898
PB - Audio Engineering Society
CY - London Uk
LA - en
AB - An algorithm for joint suppression of noise and reverberation from speech signals is presented. The method requires a handclap recording that precedes speech activity. A running kurtosis technique is applied in order to extract an estimation of the late reflections of the room impulse response from the clap while a moving average filter is employed for the noise estimation. Moreover, the excitation signal derived from the Linear Prediction (LP) analysis of the noisy speech along with the estimated power spectrum of the late reflections are used to suppress late reverberation through spectral subtraction while a Wiener filter compensates for the ambient noise. A gain magnitude regularization step is also implemented to reduce overestimation errors. Objective and subjective results show that the proposed method achieves significant speech enhancement in all tested cases.
ER -