Characteristics of Multi-Layer Perceptron Models in Enhancing Degraded Speech

Thanh Tung LE; John MASON; Tadashi KITAMURA

Characteristics of Multi-Layer Perceptron Models in Enhancing Degraded Speech

Thanh Tung LE, John MASON, Tadashi KITAMURA

Full Text Views

0

Share
Cite this

Summary :

A multi-layer perceptron (MLP) acting directly in the time-domain is applied as a speech signal enhancer, and the performance examined in the context of three common classes of degradation, namely low bit-rate CELP degradation is non-linear system degradation, additive noise, and convolution by a linear system. The investigation focuses on two topics: (i) the influence of non-linearities within the network and (ii) network topology, comparing single and multiple output structures. The objective is to examine how these characteristics influence network performance and whether this depends on the class of degradation. Experimental results show the importance of matching the enhancer to the class of degradation. In the case of the CELP coder the standard MLP with its inherently non-linear characteristics is shown to be consistently better than any equivalent linear structure (up to 3.2 dB compared with 1.6 dB SNR improvement). In contrast, when the degradation is from additive noise, a linear enhancer is always, superior.

Publication: IEICE TRANSACTIONS on Information Vol.E78-D No.6 pp.744-750

Publication Date: 1995/06/25

Publicized

Online ISSN

DOI

Type of Manuscript: Special Section PAPER (Special Issue on Spoken Language Processing)

Category

Cite this

Copy

Thanh Tung LE, John MASON, Tadashi KITAMURA, "Characteristics of Multi-Layer Perceptron Models in Enhancing Degraded Speech" in IEICE TRANSACTIONS on Information, vol. E78-D, no. 6, pp. 744-750, June 1995, doi: .
Abstract: A multi-layer perceptron (MLP) acting directly in the time-domain is applied as a speech signal enhancer, and the performance examined in the context of three common classes of degradation, namely low bit-rate CELP degradation is non-linear system degradation, additive noise, and convolution by a linear system. The investigation focuses on two topics: (i) the influence of non-linearities within the network and (ii) network topology, comparing single and multiple output structures. The objective is to examine how these characteristics influence network performance and whether this depends on the class of degradation. Experimental results show the importance of matching the enhancer to the class of degradation. In the case of the CELP coder the standard MLP with its inherently non-linear characteristics is shown to be consistently better than any equivalent linear structure (up to 3.2 dB compared with 1.6 dB SNR improvement). In contrast, when the degradation is from additive noise, a linear enhancer is always, superior.
URL: https://globals.ieice.org/en_transactions/information/10.1587/e78-d_6_744/_p

Copy

@ARTICLE{e78-d_6_744,
author={Thanh Tung LE, John MASON, Tadashi KITAMURA, },
journal={IEICE TRANSACTIONS on Information},
title={Characteristics of Multi-Layer Perceptron Models in Enhancing Degraded Speech},
year={1995},
volume={E78-D},
number={6},
pages={744-750},
abstract={A multi-layer perceptron (MLP) acting directly in the time-domain is applied as a speech signal enhancer, and the performance examined in the context of three common classes of degradation, namely low bit-rate CELP degradation is non-linear system degradation, additive noise, and convolution by a linear system. The investigation focuses on two topics: (i) the influence of non-linearities within the network and (ii) network topology, comparing single and multiple output structures. The objective is to examine how these characteristics influence network performance and whether this depends on the class of degradation. Experimental results show the importance of matching the enhancer to the class of degradation. In the case of the CELP coder the standard MLP with its inherently non-linear characteristics is shown to be consistently better than any equivalent linear structure (up to 3.2 dB compared with 1.6 dB SNR improvement). In contrast, when the degradation is from additive noise, a linear enhancer is always, superior.},
keywords={},
doi={},
ISSN={},
month={June},}

Copy

TY - JOUR
TI - Characteristics of Multi-Layer Perceptron Models in Enhancing Degraded Speech
T2 - IEICE TRANSACTIONS on Information
SP - 744
EP - 750
AU - Thanh Tung LE
AU - John MASON
AU - Tadashi KITAMURA
PY - 1995
DO -
JO - IEICE TRANSACTIONS on Information
SN -
VL - E78-D
IS - 6
JA - IEICE TRANSACTIONS on Information
Y1 - June 1995
AB - A multi-layer perceptron (MLP) acting directly in the time-domain is applied as a speech signal enhancer, and the performance examined in the context of three common classes of degradation, namely low bit-rate CELP degradation is non-linear system degradation, additive noise, and convolution by a linear system. The investigation focuses on two topics: (i) the influence of non-linearities within the network and (ii) network topology, comparing single and multiple output structures. The objective is to examine how these characteristics influence network performance and whether this depends on the class of degradation. Experimental results show the importance of matching the enhancer to the class of degradation. In the case of the CELP coder the standard MLP with its inherently non-linear characteristics is shown to be consistently better than any equivalent linear structure (up to 3.2 dB compared with 1.6 dB SNR improvement). In contrast, when the degradation is from additive noise, a linear enhancer is always, superior.
ER -