DEVICE AND METHOD FOR VOICE SYNTHESIS, PROGRAM, AND RECORDING MEDIUM

摘要

PROBLEM TO BE SOLVED: To obtain a synthesized voice which is rich in feeling by generating the synthesized voice whose voice quality is changed according to the state of feelings. SOLUTION: A parameter generation part 43 generates conversion parameters and synthesis control parameters according to state information showing the state of feelings of a pet robot. A data conversion part 44 converts frequency characteristics of phoneme piece data as voice information. A waveform generation part 42 obtains necessary phoneme piece data according to vocal sound information included in a text analysis result, processes and connects the phoneme piece data according to the vocal sound information and synthetic control parameters to generate synthesized voice data of the corresponding vocal sound and voice quality.