Unit voice editing type rule synthesizer

Abstract

Abstract PURPOSE: To smooth the tone quality or speaking to output arbitrary sentences with voice by preparing unit voices spoken with plural syllables, where phoneme circumstances are taken into consideration, with respect to voices having high degrees of coupling due to tuning of a rule synthesis device used in an audio response system. CONSTITUTION: The character string of a sentence to be synthesized is inputted from a character string input terminal 1 of a unit voice editing type rule synthesis device to a unit voice name string generation part 2 and a prosody pattern generation part 3, and the character string is converted to a unit voice name string by the generation part 2, and it is outputted to a tuning coupling degree discrimination part 4 and unit voice selection part 6. Degrees of tuning couplings between voice names are discriminated in the discrimination part 4 by the order of degrees of tuning coupling stored in a tuning coupling degree table 5 and are outputted to the selection part 6. The selection part 6 selects unit voices to be actually used in accordance with unit voice names and coupling degree discrimination results and outputs the unit voice names to a unit voice data memory 7, and the output of the memory 7 and the prosody pattern from the generation part 3 are synthesized in an editing synthesis part 8.

Inventors

• 👤 Kazuhiko Iwata
• 👤 Katsunobu Fushikida
• 👤 Masako Ichikawa

Bibliographic Details