I’ve been playing with this, but mainly get Roboto voice.
Someone said somewhere that Rap mode might be helpful, but as I’ve become more familiar with the product, it occurs to me that “Rap Mode” might only be available in specific voices that offer such a mode. Is that so? If so, any good suggestions as for which one(s)?
It also occurred to me that I might use pitch curves to provide some semblance of speech. Humans tend to (sort of) sing,or at least vary pitch (I consider the Irish brogue, which I love, to be a great example of a “sung speech” ). So far, I’ve tried just changing the notes of individual words, but that feels too drastic.
A proper speech (talk mode) would be very useful. Even for parts of a song… I have been trying to get that “non vocalist” tone == speech/talk but have not got good results from it… It needs to be built in with the engine to be able to tweak speech parameters…
The best way I’ve gotten this done is to do the speaking part myself, then use the Dreamtonics Vocoflex plugin to replace my voice with whichever model I want to be speaking instead. It worked out pretty well.
I have primarily used Vocoflex for spoken word. It has worked great or my cases. The key is to make sure you use the Vocoflex pitch shift to get your voice into the correct range for the Liam voice. The UI should have a guide that helps you find the sweet spot.