I don't usely comment on other's work because I'm a bitter person and would hate for someone to lie about my work and I feel its never good enough. So this is a first.
I can only imagine how long it took to animate that. I'm not sure which program you used to animate the voice. Either lipsink or Mimic. I'm familiar with Mimic and it appears that you let Mimic control how your model spoke. Mimic tries to sound out each vow that is in the printed word. you need to have one of your friends talk and watch how the mouth moves. You have to fool mimic as to the way the word sounds sometimes. And other times you have to go in and take out some of the breaking points of a word so the mouth moves smoothy through the word. Example The word "the" Mimic will break it up into two parts, th and eh, by doing so it will give both breaks the same the same strength, but when a person speaks the strength is on th and the eh is almost non existant and is smooth. I hope you understand what I am saying. By not adjusting the strengths it looks like it is out of sync.
I like the fact that the background is not static and you have motion, but then again in the close up of the dancers, their expressions never change. Why the close up? Which brings up another point. The lighting makes me believe you recorded the prevue and not a rendered scene. There is no active shadows, neither on the ground or the dancers. No close ups of the singer, though I can imagine that you couldn't get Mimic to work with the song even by typing in the words. Its really hard and there is a secret for doing it. I know, I spent over six months trying to figure out how to make one of my figures mimic a tersia teng song for a music video. I'm still not satisfied.
I do not know what your goal was. I don't know if you suceeded that's up to you. I did like to see what others do so I can judge my work. Maybe one day I get the courage to post it and let someone tell me what's wrong.
TimeSquare MusikClip
Re: TimeSquare MusikClip 
[