Voice compression systems for wireless telephony

Author

Bustos, Javier ; Bassi, Alessandro

Author_Institution

Dept. de Ciencias de la Comput., Chile Univ., Santiago

fYear

2001

fDate

2001

Firstpage

41

Lastpage

49

Abstract

This paper presents a comparative study between three voice compression systems for wireless telephony: CELP (code excited linear prediction), VSELP (vector sum excited linear prediction) and GSM 06. 10 (Global Standard for Mobile communications), and one system based on artificial neural networks (ANNs). The main result is that the system based on ANNs exceeds the best current performance standard (CELP). However, its speaker dependency hinders its potential standardization

Keywords

linear predictive coding; neural nets; performance evaluation; radiotelephony; source coding; standardisation; telecommunication computing; telecommunication standards; voice communication; CELP; GSM 06. 10; VSELP; artificial neural networks; code excited linear prediction; global standard; mobile communications; performance; speaker dependency; standardization; vector sum excited linear prediction; voice compression systems; wireless telephony; Artificial neural networks; GSM; Mathematical model; Nonlinear filters; Resonance; Shape; Signal generators; Speech; Standardization; Telephony;

fLanguage

English

Publisher

ieee

Conference_Titel

Computer Science Society, 2001. SCCC '01. Proceedings. XXI Internatinal Conference of the Chilean

Conference_Location

Punta Arenas

ISSN

1522-4902

Print_ISBN

0-7695-1396-4

Type

conf

DOI

10.1109/SCCC.2001.972630

Filename

972630