Description
Singing voice audio can be synthesized from score and lyric inputs using an HMM-based system. This package is useful for researchers, musicians, and voice-synthesis users experimenting with generated singing. Output quality and rights depend on the voice models, input material, and generated audio use.