Is there anyone who has any success in altering any of the Omnivocal plugin voices to sound like a particular Star..(like Roy Orbison or Elvis)
I mean I worked at it for hours trying to get lucky or stumble on something using different plugins but no changes
That is not how Omnivocal is designed. Thesame way you cannot turn your own voice into Elvis. It has two different voices at the moment and none of them sound like some celebrity.
What @Johnny_Moneto said, but there is something you can do, using third party products that change one vocal to another. Doing so is not likely to give you an Elvis or Orbison voice (certainly not legally – though there may be far-less-than-legal ways to do it). Here is what I mean, and how I used it for one specific application.
My specific Omnivocal application is using it for temporary lead vocals while working up an arrangement. This avoids my having to get a mic out to record my own vocals while things are still in flux (e.g. lyrics, tempo, etc.) while working on the instrumental side of an arrangement. I’ve done that with straight Omnivocal, typically using the male voice since I’ll eventually replace it with my own vocal. I don’t actually pay too much attention to Omnivocal settings since it is only meant to be temporary, but I do make sure the lyrics are correct (at least as of where things stand at any given moment), and I potentially do some editing on the MIDI associated with Omnivocal to work around issues it has with, for example, dropping consonants on short notes.
So far, nothing special, and only the two Omnivocal beta voices are available. However, I can render the Omnivocal instrument track to audio, disabling the Omnivocal track itself (but leaving it available in case I want to go back and change something later, redoing the steps I’m about to describe at that point). Then I can take that audio track and add the IK Multimedia ReSing extension, and change the Omnivocal voice to another voice that I like better. I’ve got ReSing MAX, so there are quite a number of voices as possibilities, so I can try those to see which one best suits my song. There are also various parameters that can be tweaked, for example for mellower or more energetic character that can be used to further alter the performance of a specific vocal model.
IK ReSing isn’t the only game in town for this. For example, Sonarworks’ SoundID AI Voice Changer can do similar things, and I know there are online services that can also do similar things. In this latter category are where some of the “far-less-than-legal” options might exist, be it directly on such sites or by finding RVC models for voices (whether made legally or not) then importing the RVC model into ReSing (I believe the Sonarworks product also uses RVC models).
By the way, modeling your own voice with RVC could be a way to import your own RVC model into ReSing and transform a rendered Omnivocal voice into your own vocal. This might actually have some merit for my own main Omnivocal use since I’ll ultimately be replacing the temporary Omnivocal track with my voice, so it would get closer than using either straight Omnivocal or a ReSing model for the temp track. I have not attempted modeling my own voice to date (and I’m not sure I will anytime soon).
One caveat on this: Any diction problems in the Omnivocal part will also influence the part transformed with ReSing. I say “influence” because there is an option in ReSing that takes its diction (i.e. phonemes) from the original RVC model instead of the source voice (ReSing in this case) or a mix of the two. But I think if Omnivocal really butchers some pronunciation, there is probably only so far a voice transformer can go in figuring out what phonemes are intended. And, if trying to model a specific voice, be it from a celebrity (probably not likely to be legal in most cases, though there may be some cases of licensed models), there is a likelihood that starting from Omnivocals phrasing, general feel, etc. will limit how close you can come to your intended voice.