Artificial Lips For Acoustics Study (and fun)

I’ve been meaning to blog about this topic for a few months now, ever since I got an email from someone asking about whether I was aware of any acoustical research projects in brass instruments using artificial lips that take into account air stream direction. As far as I know, there haven’t been any. My recollection is that the emailer was a grad student conducting research, but I’ve lost the email and my reply. If that was you (or you are similarly conducting research using artificial lips to play brass instruments), please email me again or post a comment here and tell us what you found out.

Recently I came across a couple of videos from Youtube user iSax Laboratories. This first one is a description of how they built a robot to play trombone.

And in this one we get to see and hear it in action.

He agrees with some of the commenters that it’s not a very good sound. I have to give him a lot of credit for trying this out and even if it’s not going to replace human musicians just yet, it’s a neat proof-of-concept.

Regular readers of this blog will probably already know that human musicians don’t place the mouthpiece dead center on the lips. Some of that is certainly due to the “foundation” of the teeth and gums behind the lips. However, one lip or another must predominate inside the mouthpiece and we know that the embouchure will either function as a downstream or upstream embouchure. I asked iSax Laboratories about this on his YouTube comments section and he replied that he had tried some different positions and settled on the one in the above video, where it seemed to work best.

There have been some other similar attempts. Back around 2010 Toyota built robots that apparently really played brass instruments. It supposedly blows air into the instrument and has artificial lips to produce the sound. However, I’m skeptical that the artificial lips are similar to the above robot. There has been acoustic research that uses oscillators as “artificial lips,” but I’m not certain how these Toyota robots recreate the brass embouchure. Check out the following video and look to see if you can see any artificial lips on this robot.

If the artificial lips attempt to recreate a human musician’s lips I can’t spot them on this robot. It is somewhere inside the robot, since the robot’s “mouth” seems to be simply a round hole. At least that’s what it looks like to me in this video. The resolution isn’t high enough to see any better.

Back in 2018 Uri Shaked tried to build a robot to play trumpet. You can read more about his attempt, titled “We tried to build a Robot that plays the Trumpet and Happily Failed (Sometimes Failure is The Best Option).”

On our way to Geekcon, we stopped at a local grocery store and got a plastic jar for pasta storage. As soon as we arrived the hackathon, Avi hooked it up to some lips he improvised from water-filled latex gloves, drilled a small hole in the jar, and hooked the air pump output to it, and pressed the trumpet against the lips. After a few minutes of tinkering with the position of the lips and the pressure applied by his fingers — there it was: a pure trumpet sound!

Go to the link I posted above to see more videos of their experiments, including some with robotic fingers as well.

There have been several studies done that use artificial lips to study acoustics and instrument design. As best as I can tell, the first design of using tubes filled with water was done in 1997 by J. Gilbert and J.F. Pettiot for a paper published in Proc. Institute of Acoustics, titled “Brass instruments, some theoretical and experimental results.” I haven’t read this paper, just seen references to it, so I can’t comment on it. A number of papers I have refer to their design of artificial lips as the one used to conduct additional research.

J. Wolfe, A.Z. Tarnopolsky, N.H. Fletcher, L.C.L. Hollenberg, and J. Smith published a paper titled “Some Effects of the Player’s Vocal Track and Tongue On Wind Instrument Sound” in 2003. They used two different artificial players. One used fluid filled latex “lips” that appears to be similar in design to Gilbert’s and Pettiot’s one. The other they described as, “a simple cantilever spring. We call this version of the player Phyl, for ‘PHYsicist’s Lips’.” (Wolfe, et al)

In 2007 Seona Bromage’s thesis used artificial lips made of latex rubber tubes filled with water. Bromage’s paper includes this image, which suggests that the mouthpiece was centered on the artificial lips.

Here’s a photograph of the actual “mouth.”

Bromage also compared the artificial lips playing a trombone to actual musicians playing, using a transparent mouthpiece. I have to admit that the discussion of the physics involved went over my head, so I’m not sure what to think of the results of this paper.

In fact, I’m not sure what to make of any of these acoustics papers. I *think* that I’m following the general discussion, but an awful lot of the physics are beyond my understanding. Combine that with the use of terms that mean something different to me (for example, upstream and downstream are terms that I would use to describe the general direction the air is directed as it passes the lips into the mouthpiece, but in physics they mean something completely different).

Just as musicians like me are not usually well trained in physics, I doubt that the physicists studying the acoustics of brass instrument have a well informed understanding of brass embouchure mechanics. To be honest, I don’t find many brass musicians have an accurate understanding of embouchure mechanics either. For the purposes of their physics research I guess it doesn’t make that much difference, but I am curious if modeling the lips in a more realistic way would maybe provide some insights that we could use to advance our understanding of instrument construction or brass pedagogy.

Again, if you’re engaged in research like the above, please leave a comment or drop me a line. I’d like to hear more about this and see if I can wrap my head a little better around this topic.