Strained-rough-voice conversion device, voice conversion device, voice synthesis device, voice conversion method, voice synthesis method, and program

摘要

A strained-rough-voice conversion unit (10) is included in a voice conversion device that can generate a “strained rough” voice produced in a part of a speech when speaking forcefully with excitement, nervousness, anger, or emphasis and thereby richly express vocal expression such as anger, excitement, or an animated or lively way of speaking, using voice quality change. The strained-rough-voice conversion unit (10) includes: a strained phoneme position designation unit (11) designating a phoneme to be uttered as a “strained rough” voice in a speech; and an amplitude modulation unit (14) performing modulation including periodic amplitude fluctuation on a speech waveform. The amplitude modulation unit (14) generates, according to the designation of the strained phoneme position designation unit (11), the “strained rough” voice by performing the modulation including periodic amplitude fluctuation on the part to be uttered as the “strained rough” voice, in order to generate a speech having realistic and rich expression uttering forcefully with excitement, nervousness, anger, or emphasis.

主权项

1. A strained-rough-voice conversion device comprising:
one or more processors executing: a strained phoneme position designation unit configured to designate a phoneme to be converted to a strained rough voice in a speech; and a modulation unit configured to perform amplitude modulation on a speech waveform so as to periodically fluctuate a curved outline of the speech waveform, the speech waveform expressing the phoneme designated by said strained phoneme position designation unit, wherein the amplitude modulation performed by said modulation unit on the speech waveform includes performing periodic amplitude fluctuation on the speech waveform by multiplying (i) the speech waveform expressing the phoneme designated by said strained phoneme position designation unit by (ii) a periodic fluctuation signal, the periodic fluctuation signal being generated according to a distribution of a fluctuation frequency of an amplitude envelope of a strained rough voice, the fluctuation frequency being a mode value of a frequency calculated for each of a plurality of points over a sampling period of the amplitude envelope of the strained rough voice, and the periodic fluctuation signal having one of frequencies in a range of 40 Hz to 120 Hz, and wherein the frequency of the periodic fluctuation signal is different from the fundamental frequency of the speech waveform expressing the phoneme designated by said strained phoneme position designation unit.