We were tasked with creating a lip sync video of our choice. After a little research I found a nice video to help me with my process.
The clip, seven seconds, showed a fair amount of facial movements, as well as some body. Seeing as the sequence is created from individual frames the overall character movement didn’t bother me.
Before I began I played with the supplied character model, complete with controllers and a handy to use interaction panel.
Some of the panels manipulation elements mimicked human facial movements accurately, but some didn’t. I could make the face mimic certain facial sets, like when a person is saying a particular vowel, but the other elements felt awkward to watch. Like watching someone trying to speak after a codeine injection at the dentist’s.
This limited my ability to accurately copy my chosen video clip.
On the other hand, certain facial movements within my trial sequence, I created a short sentence for the model to say, looked quite human. Knowing this I lowered my expectations for the final view of my forthcoming video sequence. Below is an example of one particular movement the tool set has to offer. It is only the up and down movement of the mouth as it opens and closes, after closing it continues upward to scrunch the top lip.
Notice the position of the mouth setting within the tool set. This allowed a small variation to manipulate the characters mouth. Moving left and right would skew the mouth in that direction. Moving the jaw position (bottom left) would stack with the mouth movement allowing the character to have a mouth that is truly agape.
By adding extra elements into each frame would allow me to create a face that would mimic a real one, though not as accurately as I would have liked.
With all of this in mind I proceeded to make my lip sync. Using the same key frame method as I did in the eight animated assets task, blogged previously to this one.
I have the video for guidance so all I need to do is copy the facial expressions straight to the respective frame.
With that said I will do some extra research into the best ways to animate a face.
Starting with some facial expressions.
At some point in our lives we have seen most, if not all, of these expressions. If anything they show how flexible the face can be. Although the movements are small, in comparison to other muscular movements, they convey so much.
They are the window to the real way we feel.
This shows a broader view of the way the face expresses our inner emotions.
\despite all of my looking I could not find a montage with a picture of real anger, maybe it is too scary for our society today. So I did a separate search for it and found one.
The anger displayed in this picture is as pure and natural as a beaming smile, just more threatening.
With the face being able to express a feeling instantly it is important to know what you are dealing with and how best to use it. Doing this effectively can negate the use of words to paint a picture. Words are slow and clumsy and often fabricated so they can be of less importance than the picture the person in front of you is displaying. On the other hand, if the words marry up with the expression they reinforce it, boost it’s meaning. The displayed emotion becomes more impressive on us and, in the occasion, you can be more sure the person is telling you the truth, this time.
The Japanese have animated emotions to fit into cartoons.
As you can see they are overstated and comical. This method allows the storyteller to use even the purest rage (4,4) without inspiring fear in the viewer ( nobody wants their children terrified of a kids tv program. This is achieved by making the expression ridiculous. In the case of the rage picture only the major details are shown (furrowed brows, narrowed eyes, and opened mouth ( cleverly substituted as a square, making the mouth into an unnatural shape implies a stiffening of the muscles controlling that area to me)), the remainder of the detail has been substituted for a reddening ( implies anger, or danger) of the face. In essence this face both shows and implies anger in equal measures, softening the actual emotion itself.
Moving on, the shapes of the mouth are the basis for being able to speak clearly.
They will be accompanied with some facial movements, like in the sets of facial expressions above. These are the focal point of the accompanying expression. All of the accompanying facial movements only happen because one of these has occurred. These movements will always happen, no matter what mood the speaker is in.
Upon looking at these I realised that it is not the shape of the mouth saying the sound, it is the shape of the sound itself. If we could not move our faces like this then we would not be able to clearly make the sound.
I found another picture that shows different sounds. These are more elongated sounds but they still have the same characteristics as the simpler vowel movements.
I also found a picture of the base mouth shapes for individual letters.
Implying there are only twelve shapes required for reciting the alphabet.
For help in understanding how to better animate a lip sync I found a site with this advice.
https://www.pluralsight.com/blog/film-games/proven-tips-animating-believable-lip-sync
Their advice is brief and to the point and I would advise any lip sync’ers to check it out.
With all of this in mind I proceeded with the lip sync. Here is my finished product.
The original video I took the lip sync from:
Although my copy of the original is not as detailed as the original I am quite happy with the results.
The sharp movements of the character as he regails his angry speech are difficult to replicate with the control rig. I am sure it can be tweaked to give a more accurate copy with the rig but I feel that a rig with more, and finer, controls would be more beneficial.
Evaluation
The lip sync was not an exciting task. Based entirely on repetition it became tiring quickly. This in part was down to the cumbersome rig. I could come to terms with it in time but at the moment it is like working with a dulled tool.
With that said animation in Maya is important, or any other animation program for that matter. I must endure with the slightly fiddly controls in the graph editor. The splines quickly overlap each other, making it difficult to effect them, post key selections. This could be addressed by Autodesk adding an isolate spline feature, allowing a much easier and tighter control over individual splines. Once this needless difficulty has been overcome I am sure I will be able to produce much better animations.
My overall performance with an animation program was satisfactory. I attained all of the basic movements and timed the sound file well ( I placed it into the video once the animation had been completed, all with no extra editing.
Practice and perfect is the direction I need to take. With a route to finding more tutorial videos to help refine my work process in this area.
My Blueprint Factory 