Let`s start applying science where we really need it. Produce a mask that can mimic a mouth singing.

They already have masks that have light panels that respond to audio frequencies. I can't see why this couldn't be modified to show different mouth shapes in response to different frequencies.