Speech Coding

Abstract

The main application of speech coding is telephony. In the traditional fixed and wireless telephone networks, speech signals are limited to the so-called telephone bandwidth, which causes the typical telephone sound of narrowband (NB) telephony. In this chapter, the authors develop a comprehensive and unified description of the most important speech coding techniques for traditional NB telephony as well as for high quality telephony. The basic concepts, such as linear predictive coding vocoder, differential pulse code modulation, and code-excited linear prediction, as well as advanced coding standards, are discussed. The second class of speech-coding algorithms comprises the parametric coders, generally called vocoders (voice coders). Vocoders provide the greatest reduction of bit rate: effectively 0.1–0.5 bit/sample can be achieved. Linear predictive coding is based on the all-pole model of speech production.

Authors 2

1. Peter Vary Aachen Institute of Communication Systems

   RWTH Aachen University

   Affiliation as printed

   RWTH Aachen University, Institute of Communication Systems, Aachen, Germany

2. Rainer Martin

   Ruhr University Bochum

   Affiliation as printed

   Ruhr-Universität Bochum, Institute of Communication Acoustics, Bochum, Germany