Applied 2: Lip Syncing Lecture

In today's morning lecture, we learnt all about lip syncing and how to do it correctly. There are several different ways to go about it, and many companies use different references, however Martin advised us to stick to vowels and a few consonants to keep the process simplistic and much easier than over complicating things. These are; A, E, I, O, U, M/V/F, L, S, K and a resting pose.

We also leant about the golden rules of lip syncing, some of which I have heard from Martin before, but it was nice to go over them again, especially as Hayley and I plan to start animating as soon as we possibly can. We learnt that light travels faster than sound, in other words your mouth moves before there is any sound coming out, so it would be unnatural for the sound to be delayed, whereas if the mouth poses are a little earlier then if anything this can improve the overall syncing. We were recommended to go perhaps 2/3 frames early but never ever late. The best way to do this for Hayley and I's project, is to make some exposure sheets and animate the frames from them. If we animate to the exact sound on the exposure sheet, then we can shuffle the sound later on, improving the overall look of our animation.

           

We then had the task of animating a piece of our audio, matched with some lip syncing. I was apprehensive to do this because after my own stop motion tests it was near enough hit and miss whether I got it right or not. However, because we had the audio in After Effects along with the different mouth poses, I was able to pair the correct ones whilst listening to the peak vowels. This made the process so much quicker and easier, but also made me realise that I really do need to print some exposure sheets because it is all about consistency and not plain estimating.

For my first test, I used way too many poses for when the actor is speaking quickly, so it is very hard to make out what he is saying, from a visual perspective. Also, when he says 'nowadays' I really struggled with trying to pair the poses with this word. I think it was because I didn't have an 'n' or a 'w' pose, so I blanked a little bit, and as you can see in this first test, it didn't really turn out well.

          

I then tried this test again, and I am much happier with it. I took some poses out with the quickly spoken part, and I also fixed the 'nowadays' section. I still think it looks a little choppy, which is perhaps because I have made 12 mouth poses for my models instead of around 6, so I am used to more choices. However this task gave me much more confidence in lip syncing and how to match it up with audio. It also made me feel less pressurised as we can move along the audio in post-production further on in the project.