Notifications
Clear all
Topic starter
21/07/2026 7:28 pm
I’m exploring Udio for a potential project: a short audio drama with two characters having a conversation across a few scenes. The big challenge for me is keeping the character voices consistent when generating each part separately.
Has anyone actually pulled this off? I’m curious about the practical workflow. Do you generate a long sample of each voice first and then keep referencing it? How does the “remix” feature hold up across multiple generations for dialogue? Any tips on prompts that help lock in a voice’s age or accent scene-to-scene would be super helpful.