Digitartic: bi-manual gestural control of articulation in performative singing synthesis

Digitartic, a system for bi-manual gestural control of VowelConsonant-Vowel performative singing synthesis is presented. This system is an extension of a real-time gesture-controlled vowel singing instrument developed in the Max MSP language. In addition to pitch, vowels and voice strength controls, Digitartic is designed for gestural control of articulation parameters, including various places and manners of articulation. The phases of articulation between two phonemes are continuously controlled and can be driven in real time without noticeable delay, at any stage of the synthetic phoneme production. Thus, as in natural singing, very accurate rhythmic patterns are produced and adapted while playing with other musicians. The instrument features two (augmented) pen tablets for controlling voice production: one is dealing with the glottal source and vowels, the second one is dealing with consonant/vowel articulation. The results show very natural consonant and vowel synthesis. Virtual choral practice confirms the effectiveness of Digitartic as an expressive musical instrument.