Seems to work fine. You add the audio file and have at least one servo assigned right ? Set your servo to have the default position and minimum set the same. I am using the SSC 32 board so I have set to 900 to max of 1600. This is all changeable of course. Once you have the audio file loaded, you then run the waveform to that servo. I will tell you that I had some issues where the waveform would only run 1/2 of the file and stop. I took two wave files and merged them together and the waveform ran the entire file. I don't have the jaw servo hooked up the skull yet, but I put an arm on the servo and held it under the jaw and it was pretty good.
What are you trying to have the relays/lights do? Typically the waveform analysis takes a vocal track and translates it 'linearly' (bad word to use) to a servo... where a relay is only ON and OFF, not sure what you'd get.
The waveform is designed to read a sound file and provide variable position output to a servo. I tried using an SSC32 relay as a device and it looks like it provided one long and delayed on signal as far as I can tell. Now you might get a dimmer to respond to the signal coming out of the servo interface, but that kind of electronics is well beyond me.