I really like how mature Eleanor sounds now. Hopefully I did her AI version justice!
(And I hope I can improve)
「いいね！」 4
Seems Eleanor is singing a lot of “Ella” these days. 
Nice cover! In some ways, it’s more natural than the Rihanna vocal.
Good job!
Sounds very comfortable, thank you for creating it~
「いいね！」 1
I tried to mimic it but then got lazy and used auto-processing (with some changes here and there)
Thanks for listening!
Nice, tight production! And an excellent showcase of Ms AI Forte’s diction.
I think that she could sound a lot more dynamic with some manual work with tension and volume – now her deadpan performance rather reveals her digitalness, if you pardon the expression.
Hi yes, I feel like you’re right on that. She does sound very monotone/deadpan-ish and digital. I’m still experimenting with tension a lot as I’m completely new to synth V and have no prior experience with vocaloid or similar products. I tried looking up tutorials but everything I could find was that it depends on each individual person and what they think sounds best
The other reason is also that Eleanor AI sounds incredibly “fuzzy” compared to her lite version. Do you have any suggestions how I could improve?
Hi!
I think that fundamental issue is that a digital voice is designed to be flexible and malleable for many different purposes. So, it is full frequency and with more or less constant intonation and minimal dynamics. It’s not an ‘MRE’ (military jargon for Meal Ready to Eat)!
So, in a given ‘performance’ the sound of the voice has to be treated as a raw recording of a real singer, which means EQ-ing, compressing, de-essing, and so on, to sculpt the basic voice according to the porject’s needs.
And then, of course, to put in a lot of work with the expressive controls of the voice, like tension, levels, pitch envelope, etc.
I have very little experience of working with artificial voices, but quite a bit of recording and mixing and editing RL voices and music… 