Greetings - an expression question...

how would you tie the up & down motion
of a ventriloquists dummy jaw to the loudness
of an audio file imported into the timeline?

eg - jaw fully shut when volume is at a minimum,
(silence)

jaw fully open when volume is at a maximum,

It should pop the jaw open and shut, in time with
thw waveform, effectively automating the "lip sync"
to whatever audio track is imported?

What would the expression syntax be?

Dan