A graph-based UI in Voicelab would allow users to split their audio signal, run it through different effects simultaneously, and mix them back together.