I'm learning as much as possible about Motion Live before investing. I've seen something I cannot understand.

Evidently it is possible, in fact it is the default, to record facial movement WITHOUT AUDIO. Uh, can somebody please tell me why ever I might want to do that? The answer may be perfectly obvious, but I'm missing it.

Thanks.


"Less clicks good, more clicks bad."

Done it a million times -- for me, it's often the default (more so than ever now with Acculips).

Let's talk pre-Acculips - in the past I often did lip-sync work, syncing to an audio file someone else had done (either my own actors or something off the web).  Post Acculips it's the ONLY way to use Motion Live, IMHO, as you will use Acculips to generate your sync from that audio file, and then use Motion Live to "finish it off" with the appropriate facial expressions (I believe RL has done at least a couple of tutorials on this approach).

Now, there's nothing wrong per se with recording the audio at the same time.  It CAN be used with Acculips post recording (although, again, I much prefer to add the expressions afterwards).  But I almost never do it anymore.

---

Alienware Aurora R12, Win 10, i9-119000KF, 3.5GHz CPU, 128GB RAM, RTX 3090 (24GB), Samsung 960 Pro 4TB M-2 SSD, TB+ Disk space
Mike "ex-genius" Kelley

Thank you Mike.
I'm hoping (expecting, even) that the first pass lipsync is good enough to use on all but mouth closeups.
Is this realistic? My inner engineer says, avoid fiddling.
---------
I can see I'll need to try this out to comprehend it.


"Less clicks good, more clicks bad."

Acculips is better than Live Face at lipsync.  It really is.  And it's kind of an "either-or" situation (not exactly, but somewhat).  Acculips creates visemes and Live Face creates expressions, and they are two different things.  They can (and are) blended, but if you are going to use Acculips you really don't want to use Live Face for lip sync on top of that.  What you want to do is to use the expressions to make a character smile, or blink, or express through their eyes, etc.

The Acculips viseme track is useful in other ways that Live Face expressions for Lip Sync cannot be used.  For example, after the fact you can adjust the visemes to reflect the different ways people speak, so mumbling, shouting, etc.  You can also load in various types of speech - a Scotsman, for example, or some other you've defined.  In short, the ability to edit the lips in this manner is AMAZINGLY powerful and nothing you can come close to with recording those lips at the same time.

Again, you can take that audio you record and push it through Acculips to get the visemes, but I don't think this is the best workflow (in that case you'd mask out the lips for your live face but I don't like this approach because then smiling and frowning and such can't be done).  Trust me, once you've done the Acculips/Live Face overlaid on top you won't want to do it any other way.

---

Alienware Aurora R12, Win 10, i9-119000KF, 3.5GHz CPU, 128GB RAM, RTX 3090 (24GB), Samsung 960 Pro 4TB M-2 SSD, TB+ Disk space
Mike "ex-genius" Kelley

"Acculips creates visemes and Live Face creates expressions, and they are two different things."

Oh... I'm glad you warned me, I did not expect this.
This will take some fooling around.
...............................
I wonder if face motion and lipsync can be created in another program, then imported as an FBX. The audio track would then be added manually, I'm guessing.
Would that be even remotely feasible?


"Less clicks good, more clicks bad."

Acculips uses a proprietary process that Apple controls.  It's based on morphs and you could not do this in any outside program (you can't actually do facial animation and import it from any place else anyway).

---

Alienware Aurora R12, Win 10, i9-119000KF, 3.5GHz CPU, 128GB RAM, RTX 3090 (24GB), Samsung 960 Pro 4TB M-2 SSD, TB+ Disk space
Mike "ex-genius" Kelley

OK... Now that a few days have gone by, I can grasp what you said. Sort of.
I "Acculips" the voice first. Now it's in the timeline.
Now, using Live Face, I add in the facial expressions. Right?
That means I have to match the facial expressions to the voice track. That sounds pretty hard to do. Do you not have to memorize the voice track before recording the facial movements?
I can see it working with a short sentence, but the longer the script gets, the more difficult it would be to synchronize the two -- no?
What am I missing?
I can record both at the same time, then afterwards apply Acculips to the voice, it seems to work well. Is it a matter of being able to edit the voice track separately, outside of iClone?


"Less clicks good, more clicks bad."

Hard to act?  I suppose if you're not an actor.

As someone trained in the theater doing the same part over and over (and over) again is easy-peasy.  But it's also the ONLY way to do things with pre-recorded audio (often the case with remote talent).  So, no, nothing to it and you'll see it time and again in the RL videos (plus you can course correct easily).  Hell, you HAVE to do this with body motions (most of the time you post record the body in sync to the audio track, so you get your mocap that way.  iClone doesn't have the horsepower to do it all at the same time, although there are many who try.  And body motions are MUCH more important than facial motions most of the time).

But to each his own - you develop your own workflow and shouldn't really care what anyone else does (I mean, it's all right to take some advice but we ALL end up doing it the way we want to in the end).

---

Alienware Aurora R12, Win 10, i9-119000KF, 3.5GHz CPU, 128GB RAM, RTX 3090 (24GB), Samsung 960 Pro 4TB M-2 SSD, TB+ Disk space
Mike "ex-genius" Kelley

Post Quoted Reply
Would you please tell me *why* it is desirable to enter the data in two steps rather than one? I am in sincere earnest.


"Less clicks good, more clicks bad."

I just told you above - I'm not sure I can make it any clearer.

But you will find your own way.  Just be aware that time is nature's way of keeping everything from happening all at once <g>.

---

Alienware Aurora R12, Win 10, i9-119000KF, 3.5GHz CPU, 128GB RAM, RTX 3090 (24GB), Samsung 960 Pro 4TB M-2 SSD, TB+ Disk space
Mike "ex-genius" Kelley