A microcomputer-based speech-output system for generating automated weather reports

Author

Eady, Stephen J. ; Dickson, B. Craig ; Snell, Roy C. ; Woolsey, J.R. ; Ollek, Parm

Author_Institution

Speech Technol. Res. Ltd., Victoria, BC, Canada

fYear

1991

fDate

9-10 May 1991

Firstpage

473

Abstract

The authors describe a microcomputer-based speech-concatenation system that is being developed to produce automated voice output for the broadcast of weather forecasts. The system consists of an inventory of prerecorded speech units (words and phrases of English), and the software that is necessary to translate a weather forecast from a text format into an audio format. Speech units are processed and encoded using residual-excited linear predictive coding. This method allows changes to be made to certain speech parameters without significant degradation in speech quality. Such changes are necessary to ensure that the voice output for weather broadcasts sounds as natural as possible

Keywords

broadcasting; encoding; filtering and prediction theory; microcomputer applications; speech analysis and processing; telecommunications computing; voice equipment; weather forecasting; English; audio format; automated weather reports; inventory; microcomputer; phrases; prerecorded speech units; residual-excited linear predictive coding; software; speech parameters; speech processing; speech quality; speech-concatenation system; speech-output system; text format; voice output; weather broadcasts; words; Application software; Broadcasting; Concatenated codes; Linear predictive coding; Radio transmitters; Speech; Telephony; Timing; Weather forecasting; Wool;

fLanguage

English

Publisher

ieee

Conference_Titel

Communications, Computers and Signal Processing, 1991., IEEE Pacific Rim Conference on

Conference_Location

Victoria, BC

Print_ISBN

0-87942-638-1

Type

conf

DOI

10.1109/PACRIM.1991.160779

Filename

160779