Ha. Voice synthesizers and TTS systems (and NLP in general - dead electronics imitating this very intimately human thing, speech and language) always fascinated me, so far that this was a significant reason for me to study CS and computational linguistics.

This is literally some of the impossible sci-fi tech I dreamt of as an undergrad. Crazy. I'm still a bit in disbelief how fast things currently move on this front.

Interestingly, suno.ai is also able to imitate the very robotic and staccato-like intonation of Vocaloids: https://app.suno.ai/song/f43e9c46-92d3-4171-bdd9-026213d6772... - everything comes around. :)