I'm still wondering if, creating an UST, you can base it according to your interpretation of the song, to make it more human e.g. If I sing matrioshka and I record it with background noise reduction, could I (not me obviously because I don't use UTAU) create the UST? This could be efficent in every song, but especially in songs that need something more... I mean... expressive. I don't consider just, for example, matrioshka sang by me, but maybe people could copy/paste the pitch... from other versions (utaute ecc.)