Síntese de voz com qualidade.

Full text
Author(s):	Evandro Bacci Gouvêa Total Authors: 1
Document type:	Master's Dissertation
Press:	São Paulo.
Institution:	Universidade de São Paulo (USP). Escola Politécnica (EP/BC)
Defense date:	1993-12-17
Examining board members:	Geraldo Lino de Campos; Normonds Alens; Demi Getschko
Advisor:	Geraldo Lino de Campos
Abstract
A phoneme-based speech synthesis system was implemented. The basic idea bearing this system was the construction of a library of phonetic units, which could be phonemes, diphones, triphones etc., and the concatenation of those basic units to build a spoken string. The phoneme\'s representation in the library was based on a linear prediction model. The filter which models the vocal tract was represented by Line Spectrum Pairs (LSP). The excitation was represented by the code excited linear prediction (CELP) algorithm. After the library was constructed, algorithms for smoothing the parameters transitions were developed, seeking a better quality of the synthesized speech signal. (AU)