This paper presents the design of a companding nonuniform optimal scalar quantizer for speech coding. The quantizer is designed using two neural networks to perform the nonlinear transformation. These neural networks are used in the front and back ends of a uniform quantizer. Two approaches are presented in this paper namely adaptive critic designs and particle swarm optimization, aiming to maximize the signal-to-noise ratio. The comparison of these optimal quantizer designs over a bit-rate range of 3-6 is presented. The perceptual quality of the coding is evaluated by the International Telecommunication Union's Perceptual Evaluation of Speech Quality standard


Electrical and Computer Engineering

Keywords and Phrases

ACDs; Adaptive Critic Designs; International Telecommunication Union; PESQ; PSO; Perceptual Evaluation of Speech Quality Standard; Adaptive Critics; Neural Nets; Neural Networks; Nonlinear Transformation; Nonuniform Optimal Scalar Quantizer Designs; Particle Swarm Optimization; Perceptual Evaluation of Speech Quality; Quantization; Signal-To-Noise Ratio; Speech Coding

International Standard Serial Number (ISSN)


Document Type

Article - Journal

Document Version

Final Version

File Type





© 2007 Institute of Electrical and Electronics Engineers (IEEE), All rights reserved.

Full Text Link