The KTH Rule System for Singing Synthesis

This article contains a description of rules controlling the singing synthesis at the Department of Speech Communication and Music Acoustics at the Royal Institute of Technology (Swedish Royal Institute of Technology-KTH) in Stockholm. The synthesis of singing has been important in our research for a long time. The rules controlling the singing synthesizer MUSSE DIG are implemented in a programming environment originally developed for a text-to-speech system. There are context-dependent rules for pronunciation of vowels and consonants, as well as rules for musical performance. The latter rules create crescendi, tempo, and vibrato changes, etc., depending on the musical context as defined by a score file. The rules were developed using an analysis-by-synthesis strategy, i.e., vocal performances are synthesized, the result is analyzed, and then the rules that control the synthesis are accordingly improved. In this article, musical rules, and general rules for consonants, vowels, and some special singing techniques are described.