Import Audio
In this lesson, we mainly talk about how to import audio files into Maya and follow the audio to do mouth animation. Only audio with WAV and AIFF format can be imported into Maya, and you only need to drag directly to the timeline to display the green audio track.
The next step is to get familiar with the models. In class, the teacher showed some models, and their mouth binding is complex. That is to say, the mouth shape corresponding to the corresponding letter can be found in the mouth attribute, which will save a lot of time.
Because I don’t have a professional knowledge of English pronunciation, I found some 2d and 3d references to help me understand and be familiar with this knowledge.
Reference
Then I recorded a video for reference by myself, in order to facilitate the phonemes animation.
Key points
1. According to the shape of the mouth and the position of the tongue. Pay attention not only to the upper and lower lips of the mouth, but also to the teeth.
2. English is a rhythmic language. In addition to using lips, teeth and tongue as the basis of pronunciation, the emphasis of notional words and the weakness of functional words produce the rhythm of light and heavy. Then, according to the context, personality and emotion, we should emphasize or weaken the pronunciation.
Animation
In this process, I will focus on the key frames of two parts, one is the mouth on the face AI and its attribute bar, and the second is the rotation axis y of the controller that controls the opening and closing of the mouth. At the beginning, I would adjust the shape of my mouth to the corresponding letter pronunciation, but I found that it was very easy to not match because of the fast speed. At this time, I found that I should first close the key frame of the mouth to determine the opening and closing of the mouth, so that I can determine which frame to adjust the mouth shape.
In fact, when speaking at a relatively fast speed, the amplitude of the mouth won’t be exaggerated. Because of the lack of time, I didn’t make every mouth shape very obvious and exaggerated. Instead, I tried to make sure that I could listen to the audio without making it strange. Sometimes the pronunciation needs to be superimposed by two letters at the same time. This is not done all at once, but some changes are added in the process of repeated proofreading.
This process is much smoother, but since the speed is too fast, much mouth shape can’t be seen clearly. At this time, I didn’t realize that the reason why my movements are always not smooth is that the mouth shape is too large. I always thought that the mouth shape is not big enough and it is not obvious. In fact, when I saw some tutorials later, I realized that I had done too many mouth shape changes in too short a time, leading to some not smooth.
There’s a lot of tongue licking.
In fact, I found that because the speed of this sentence is very fast, I hope its mouth shape can match the audio more accurately. If I choose these points and move a frame or two, it will look much more comfortable.
Phonemes
Tongue
This is all phonemes animation.
When I have finished all the mouth animation, the next step I hope is if some facial expressions, including blinking, make him look less rigid. It’s more difficult than I thought, because my model is not smooth when adding other actions. I can’t watch the changes of actions in real time, which makes my work very time-consuming.
In this process, I basically joined the shaking of my head, the slight rise of facial muscles and eyebrows, and several eyebrow picking movements. In fact, I’m not very satisfied with these expressions, but I don’t have much animation skills about the expressions, so I will modify them if I learn the skills later . For the time being, it’s just to match the mouth shape with the audio and add some simple head manipulation.
Eyebrow picking
Facial muscles
Nose
Head movement
Blink
Then I found some tutorials about adjusting the lips.
- Try not to make all the facial features move together, nor too symmetrical, that is, there are some wrong frames in the nose, mouth and muscles, which is also part of the overlapping movement.
- Some exaggerated mouth patterns should be neglected to ensure the continuity of the movements.When speaking at a high speed, some exaggerated mouth patterns should be properly ignored to ensure the continuity of actions.
- When speaking at a slow speed, the mouth shape can be exaggerated, especially some obvious pronunciation.
This is my final animation for the current, and I think I’ll probably spend some time later to modify my animation with regard to the tutorials.