Just to chime in with a quick note here on one of the methods we use in our studio...

Some audio software will allow you to put markers in your audio file. I think quite a few allow that. Anyway, we actually use a few different audio applications, but all of them have some of functionality that allows us to place these markers in the audio file and then export them as a txt file. One can open the text file and see the time of each marker, the marker name and any notes you may have entered there. The time is displayed based on the timecode settings in the software, so you can work with either actual SMTP timecode or absolute frames. Absolute frames is what I prefer for most work. What this does, then, is allow one to listen to the audio, dropping markers and placing notes in them for whatever animation needs to happen at that point, then save all that out as txt file which then acts as, basically, an old fashioned "dope-sheet" or event list. Even when using automated lipsync applications, I still find this to be a very handy bit of setup work as it allows you to really plan out the animation and think through things in the performance that automation just can't nail down. When using this technique for straight forward actions, it allows one to have an immediate, textual reference for the timing of your movements, so even if for some reason you don't want the audio loaded while you're animating, you can still reference your event list and be quite accurate with your work.

-Les