Select two voice models, set your desired blend percentage, and blend them into an entirely new voice for your personal library.