new voice(s), based on the text to speeches you all hopefully know, (like https://www.tetyys.com/SAPI4/ or https://discordier.github.io/sam/ or even the ones you see in FL studio), it can both be its own voice or be used in voicelab (for if the character you want has an effect in its voice, e.g. more louder, reverb, different pitch, etc)

With options (if to be real) COULD include

pitch (in case the different pitch part isn't your fancy),

what voice,

choir,

wait (Waits then speaks what you said in tts) or "mix as speak" (speaks while you are, trying to find out what are you saying while its being said)

and MAYBE match tone (it, as stated in the title, tries to match your pitch/tone when it repeats (e.g. if you wanna add emotion or sing)