for what reason(s) do you feel the need to run a sequencer off an audio signal?  

I'm very much willing to bet that all of them will end up being totally and easily do-able using message objects