Mandarin Singing Voice Synthesis Using an HNM Based Scheme

In this paper, HNM (harmonic plus noise model) is enhanced and used to design a scheme for synthesizing Mandarin singing voice. Enhancements made include synthesizing signals with higher fluency level and keeping the timbre of synthetic singing voices consistent. In terms of the signal synthesis equations rewritten here, a Mandarin singing voice synthesis system is constructed, in which each Mandarin syllable is recorded only once. This system can parse a song score file and synthesize its lyric syllables' signals in real-time. Besides, the skill of portamento singing is realized. According to perceiving the synthetic songs, the timbre is indeed consistent, and the signals are very clear and natural.