Relatively easy to train your own voice, but you'll need 2 applications, Applio and w-okada. You need good source material and a modern GPU. Info in the discord. Come through for details. codyhurst.com

Yeah pretty much what he said but also do note. The source material needs to be of at least moderate quality. The more accuracy you want, the more material you need. If you can get a wide range of vocal patterns and expressions in the audio it will give you the best results

no you need consistant one though, so make microphone it same one, or you get very bad quality like electronic sound high pitched