I'm recently working on making a version of Ubi Caritas (Durufle) through a software called Vocaloid.
This is just the first few seconds: http://pds16.egloos.com/pds/201004/08/68/test_ubc.mp3
(File plays after 7 seconds)
What do you think? It's nowhere near as good as actual human interpretation, but any tips in making this more human-like?
Thanks!
Lol, the vowel /u/ is pronounced as in Japanese
But in general it's an interesting result. Is there a way to configure Vocaloid to maintain intensity for a longer period and avoid this "wow effect"?
Thanks for your comment, Carlos! But I have trouble understanding your advice; could you specify what you mean by "intensity" and "wow effect?" Do you mean that I should reduce the vibration in longer sounds?