Lipsync in Maya
Author
I was wondering if its possible to construct lipsyncing for animated characters with maya in a similar fashion to this:
1) With the animated head available, set some "move-points" on certain areas around and on the lips (for simplicity, lets say left mouth corner is move-point A, right corner B, upper mid lip C, and lower lip mid D).
2) Shape the mouth for each phonetic sound thats necessary to form every word youll need, and store the x- and y-coordinates (z might be necessary too, but lets leave it out for now) for each of the move-points A-D in an array.
Move-point A for phonetic sound "a": x=2, y=3
Move-point B for phonetic sound "a": x=6, y=3
Move-point C for phonetic sound "a": x=4, y=4
Move-point D for phonetic sound "a": x=4, y=2
Move-point A for phonetic sound "m": x=1, y=3
Move-point B for phonetic sound "m": x=7, y=3
Move-point C for phonetic sound "m": x=4, y=4
Move-point D for phonetic sound "m": x=4, y=3
3) Then code phonetic sentences (for instance, "mama" will infact be "m" -> "a" -> "m" -> "a") so that the move-points will move accordingly to the phonetic sound that is being called.
So, is this a (relatively) easy thing to do in Maya using MEL?
Thanks
1) Your "move-points" would probably be Clusters. You can just group a number of vertices or CVs into a Cluster, and move it around freely.
2) For this, I could only imagine the use of Set Driven Keys. You might make a Locator near the mouth that uses numerous attributes you label "m", "a", etc to change the x,y,z coordinates of the Clusters. Set Driven Keys can seem fairly complicated to set up if you aren''t familiar with them, so find a good tutorial.
3) I don''t know too much of MEL, but I''m sure it would be possible to automate the Set Driven Key process. However, I''m certain it''d be a bad idea. A golden rule is: Never let the computer animate. Your word structure would lack all sense of timing. For this...stop thinking like a programmer, and start being an animator.
....also, you might want to consider Blend Shapes. If you''re not familair with them, they allow you to make multiple copies of the head, and then you can model the seperate heads to form all the mouth shapes you need. Then using Blend Shapes, you then "blend" them all together, and using slider bars, can make the original head incrementally take on the shape of all the others. You can also mix and match the various heads to get new results.
2) For this, I could only imagine the use of Set Driven Keys. You might make a Locator near the mouth that uses numerous attributes you label "m", "a", etc to change the x,y,z coordinates of the Clusters. Set Driven Keys can seem fairly complicated to set up if you aren''t familiar with them, so find a good tutorial.
3) I don''t know too much of MEL, but I''m sure it would be possible to automate the Set Driven Key process. However, I''m certain it''d be a bad idea. A golden rule is: Never let the computer animate. Your word structure would lack all sense of timing. For this...stop thinking like a programmer, and start being an animator.
....also, you might want to consider Blend Shapes. If you''re not familair with them, they allow you to make multiple copies of the head, and then you can model the seperate heads to form all the mouth shapes you need. Then using Blend Shapes, you then "blend" them all together, and using slider bars, can make the original head incrementally take on the shape of all the others. You can also mix and match the various heads to get new results.
This topic is closed to new replies.
Advertisement
Popular Topics
Advertisement
