Abstract
This paper shows that a new Data-to-Speech system, called GoalGetter, could be created relatively easy from an existing one, the Dial-Your-Disc (DYD) system, despite the fact that the domain, the language and the speech output technique differ. It is shown that the original system is easily portable and that problems that arose after porting can receive generalized solutions so that the system as a whole is enhanced. The speech output technique used ( specific variant of the phrase concatenation technique) finds an optimum between the requirements of high quality output and flexibility. In addition, requirements for setting up a phrase database necessary for this technique are specified.
Original language | English |
---|---|
Pages (from-to) | 66-75 |
Number of pages | 10 |
Journal | IPO Annual Progress Report |
Volume | 31 |
Publication status | Published - 1996 |