Home // COMPUTATION TOOLS 2018, The Ninth International Conference on Computational Logics, Algebras, Programming, Tools, and Benchmarking // View article


A Method for Recovering Speech Signals Heavily Masked by Music Based on the Affine Projection Algorithm

Authors:
Robert Alexandru Dobre
Constantin Paleologu
Cristian Negrescu
Dumitru Stanomir

Keywords: multimedia forensic; noise reduction; adaptive filtering; affine projection algorithm

Abstract:
The importance of multimedia materials in justice is increasing. For example, a security camera recording could provide the evidence needed to clarify a given situation. The problems that arise are linked to the authenticity or intelligibility of the materials. There are situations in which the key material, (for example, a dialogue) is heavily masked. This paper presents the performances obtained by the Affine Projection Algorithm within a method for recovering speech signals masked by music. The results help in deciding if audio monitoring a certain acoustic environment could prove useful if the proposed method for extracting the speech is used afterwards.

Pages: 29 to 34

Copyright: Copyright (c) IARIA, 2018

Publication date: February 18, 2018

Published in: conference

ISSN: 2308-4170

ISBN: 978-1-61208-613-2

Location: Barcelona, Spain

Dates: from February 18, 2018 to February 22, 2018