A modified Armijo rule for the online selection of learning rate of the LMS algorithm

Autor: Lazaros Polymenakos, Danilo P. Mandic, Christos Boukis, Anthony G. Constantinides

Publikováno v: Digital Signal Processing. 20:630-639

The use of the Armijo rule for the automatic selection of the step size within the class of stochastic gradient descent algorithms is investigated, and the Armijo rule learning rate least mean square (ALR-LMS) algorithm is introduced. This algorithm

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::a2b2b47aac79cd6fd6419653115395ff
https://doi.org/10.1016/j.dsp.2009.09.003

Toward bias minimization in acoustic feedback cancellation systems

Autor: Christos Boukis, Danilo P. Mandic, Anthony G. Constantinides

Publikováno v: The Journal of the Acoustical Society of America. 121:1529-1537

A novel technique for bias suppression within acoustic feedback cancellation systems is proposed. This is achieved based on the use of all-pass filters in the forward part of the hearing aid. The poles of these filters are made time-varying, which re

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::014d7595a303daa950dbc3db8c6cb2eb
https://doi.org/10.1121/1.2431341

Cascaded dynamic noise reduction utilizing VAD to improve residual suppression

Autor: Fotios Talantzis, Theodoros Petsatodis, Lazaros Polymenakos, Christos Boukis

Publikováno v: Petsatodis, T, Boukis, C, Talantzis, F & Polymenakos, L 2013, Cascaded dynamic noise reduction utilizing VAD to improve residual suppression . in 2013 18th International Conference on Digital Signal Processing, DSP 2013 ., 6622696, IEEE Press, International Conference on Digital Signal Processing proceedings, pp. 1-6, International conference on Digital Signal Processing, Fira, Greece, 01/07/2013 . https://doi.org/10.1109/ICDSP.2013.6622696
DSP

Speech processing systems often operate in noisy and reverberant environments. Their operation is subject to the accuracy of the underlying noise reduction algorithm, that aims to reduce noise present in the signals that are captured by the employed

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::bb39571316d10c444b108efb6c382dc0
https://vbn.aau.dk/da/publications/0597eb9c-af9c-4b5a-87ec-f9e811342bb6

Empirical mode decomposition VAD based on multiple sensor LRT

Autor: Theodoros Petsatodis, Christos Boukis, Fotios Talantzis

Publikováno v: 2013 Constantinides International Workshop on Signal Processing (CIWSP 2013).

Voice Activity Detection (VAD) remains a challenging task given its dependence on adverse noise and reverberation conditions. The problem becomes even more difficult when the microphones used to detect speech reside far from the speaker. In this pape

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::cc330d6ac0b21a343000bb8f83571bc8
https://doi.org/10.1049/ic.2013.0009

Convex Combination of Multiple Statistical Models with Application to VAD

Autor: Fotios Talantzis, Theodoros Petsatodis, Ramjee Prasad, Christos Boukis, Zheng-Hua Tan

Publikováno v: Petsatodis, T, Boukis, C, Talantzis, F, Tan, Z-H & Prasad, R 2011, ' Convex Combination of Multiple Statistical Models with Application to VAD ', I E E E Transactions on Audio, Speech and Language Processing, vol. 19, no. 8, pp. 2314-2327 . https://doi.org/10.1109/TASL.2011.2131131
IEEE Transactions on Audio, Speech, and Language Processing

This paper proposes a robust voice activity detector (VAD) based on the observation that the distribution of speech captured with far-field microphones is highly varying, depending on the noise and reverberation conditions. The proposed VAD employs a

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::65819629586327acd6ffe3db7d4ccf27
https://vbn.aau.dk/da/publications/7c386252-54a7-492d-8e05-5d31ba630ddb