The paper discusses the challenge posed by nonlinear distortions in preserving the quality of audio and speech signals. This research involves a comprehensive analysis to determine the optimal approach for Nonlinear Acoustic Echo Cancellation (NAEC) and audio signal processing. The experimental results are evaluated not only in terms of signal quality but also in relation to its intelligibility. To tackle this issue, nonlinear models are employed, and spline-based estimation has drawn attention from the scientific community due to its promising performance in system identification and various tasks. We propose a novel framework centered around a Functional Link Adaptive Filter (FLAF), designed for different classes of nonlinear systems. This framework improves the performance consistency by incorporating a Functional Expansion Block (FEB) before the spline nonlinearity. Our simulations demonstrate convincing results that outperform traditional FLAF models.

Spline adaptive exponential functional link filter for nonlinear acoustic echo cancellation / Nezamdoust, A.; Scarpiniti, M.; Uncini, A.; Comminiello, D.. - (2024), pp. 216-220. (Intervento presentato al convegno 32nd European Signal Processing Conference, EUSIPCO 2024 tenutosi a Lyon; France).

Spline adaptive exponential functional link filter for nonlinear acoustic echo cancellation

Nezamdoust A.;Scarpiniti M.;Uncini A.;Comminiello D.
2024

Abstract

The paper discusses the challenge posed by nonlinear distortions in preserving the quality of audio and speech signals. This research involves a comprehensive analysis to determine the optimal approach for Nonlinear Acoustic Echo Cancellation (NAEC) and audio signal processing. The experimental results are evaluated not only in terms of signal quality but also in relation to its intelligibility. To tackle this issue, nonlinear models are employed, and spline-based estimation has drawn attention from the scientific community due to its promising performance in system identification and various tasks. We propose a novel framework centered around a Functional Link Adaptive Filter (FLAF), designed for different classes of nonlinear systems. This framework improves the performance consistency by incorporating a Functional Expansion Block (FEB) before the spline nonlinearity. Our simulations demonstrate convincing results that outperform traditional FLAF models.
2024
32nd European Signal Processing Conference, EUSIPCO 2024
acoustic echo cancellation; functional links; nonlinear adaptive filters; nonlinear modeling; spline adaptive filters
04 Pubblicazione in atti di convegno::04b Atto di convegno in volume
Spline adaptive exponential functional link filter for nonlinear acoustic echo cancellation / Nezamdoust, A.; Scarpiniti, M.; Uncini, A.; Comminiello, D.. - (2024), pp. 216-220. (Intervento presentato al convegno 32nd European Signal Processing Conference, EUSIPCO 2024 tenutosi a Lyon; France).
File allegati a questo prodotto
File Dimensione Formato  
Nezamdoust_Spline_2024.pdf

solo gestori archivio

Note: Paper EUSIPCO 2024
Tipologia: Documento in Post-print (versione successiva alla peer review e accettata per la pubblicazione)
Licenza: Tutti i diritti riservati (All rights reserved)
Dimensione 946.4 kB
Formato Adobe PDF
946.4 kB Adobe PDF   Contatta l'autore

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11573/1726627
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 0
  • ???jsp.display-item.citation.isi??? ND
social impact