teal hawk
Hi, guys. Let's say for example I want to make a cover, the original song is English and the desired is Spanish. Or it's a situation English - English with changed lyrics.
I use RVC method, but my problem is while it keeps the original singer voice, it doesn't keep the way the original singer sings that songs. Unfortunately, it mimics the singing process of audio sample(of Spanish singer)
So, what I am looking for is some kind of abomination(mix), where the Spanish singer's audio sample provides for an AI example how to pronounce Spanish words, while English sample is responsible for singer's voice and singing process.