Slashdot Mirror
Clarinet Wins Robotic Orchestra Competition
Sasha writes "The Australian designed robotic clarinet beat out Dutch and Finnish entries this year at the robotic orchestra competition. The researchers don't expect to replace human musicians, but are instead interested in what makes the difference between playing music well and playing music poorly. There is also a video available of the performance."
I, for one, welcome our new clarinet playing overlords.
Actually, the answer is yes.
Doomoo ari-datou.. Missu-ta Roh-baa-tou..
Previously: "Linux... Toward the Sunrise..." Now: "Linux... Toward the-- No, now, part of Every Sunrise"
Look on the bright side. They could have made a robotic bagpipe player.
When our name is on the back of your car, we're behind you all the way!
Captured by robots has a whole band of robots that play their own instruments.
Can it play 92 cents below the lowest octave of E-Flat?
Karma: SELECT `karma` FROM `users` WHERE `userid`=138474;
A video of the second place winning guitar picking robot from the Netherlands. Sounds good to me. :)
After the performance, the clarinet was overheard saying, "All your brass are belong to us."
-g.
Well, I prefer this one :-)
http://vimeo.com/1109226
Oolite: Elite-like game. For Mac, Linux and Windows
Here is the contest website and a longer article.
Commenting without saying anything is a much bigger weakness in this community. Well followed.
Nerd rage is the funniest rage.
Are you kidding me? It is what humans do best, and I would say it is what separates humans from basically every other animal in the world we know of. We have the ability to spend resources on problems that aren't really useful to us yet, but one day they might be. All to play have fun and do something that wasn't possible before. Because of our natural playful attitudes towards life, we have been able to do very useful things. Using your criterion, some of man's greatest achievements are not worthwhile. Going to the moon was nothing more than an ego trip, but the knowledge gathered from it may one day save our species. Who knows. I do know that playing with things is how we learn, and learning in itself is worthwhile.
When someone says, "Any fool can see
I, for one, welcome our robot-played overture.
Knowledge is how to play a game, intelligence is how to win, wisdom is knowing what game to play.
I played clarinet for a dozen years - it's not very fair in some ways, simply because the clarinet is one of the most basic instruments in an orchestra, where a musician has the least influence on the actual sound/tonal quality beyond a certain level. Even some percussion instruments give you more ability to influence the sound of a note.
-- I really need to bleed off some of this
Live today, because you never know what tomorrow brings
Same could be said of music, art, etc... doesn't make it any less interesting or important to people. Curiosity, tinkering, and "because I can" have lead to all sorts of amazing things. Just because you don't understand the motivation doesn't mean it's not worthwhile to someone in some way.
You are using English. Please learn the difference between loose and lose; they're, there, and their; your and you're.
Just because you don't see the value or purpose of a line of research does not mean it doesn't have value. At the end of the day, what we have is more in depth knowledge about what makes music good and bad, how to coordinate fine motion in robots, and numerous other advances.
Just so you know, you come across as a supercilious ass in this post. All you've shown is your ignorance regarding science and research, and your desire to tell other people how to spend their time. If that's not the image you wish to project to others, I suggest you work on your social skills.
- None can love freedom heartily, but good men; the rest love not freedom, but license. -- John Milton
And robotics is cool, fine, maybe not everyone thinks so, I do. Its partly while im building one. WHY was it undertaken? No good reason in the grand scheme of things, but it amuses me and furthers my understanding of the subject, and things i learn here might have, shock horror, real world uses. In fact, i think you'd struggle to explain how it doesn't.
If we didn't bother to try new things without an immediate and definite gain for society we'd likely still be sat around in caves wearing animal skins.
"I may be full of crap about this game, and I may be wrong, and that's fine." -Jack Thompson
This blows.
We used to have a Bill of Rights. Now, with the rights gone, all we have left is the bill.
I would take your comment even further, and say that in general "fate" tends to laugh at plans, and (anecdotally, of course) most efforts undertaken to advance along a set path and accomplish a specific goal are fraught with failures and setbacks. On the other hand, efforts undertaken for the pleasure of doing them frequently not only yield our best culture, but our most innovative advancements (and at worst, they were generally at least amusing).
That's not to say we shouldn't set goals, we should just expect that our true successes in life will come from what others might view as frivolous.
If I had a nickel for every time I had a nickel, I'd be richcursive!
BobMcD, I concur that basically the clarinet really isn't being played very well. Fundamentally, music can be modeled/thought-of as a highly-coordinated set of stimuli that act on a listener in a very wide variety of ways (see www.omsmodel.com for some specifics - over 100 high-level kinds of stimulation). When music is performed/realized just as a sequence of notes (e.g. this clarinet performance), the overall stimulation is rather low and boring.
You are making a false assumption that the only thing that comes out of the project is a robot that plays the clarinet. There are several other outputs:
First, some students learned how to build a robotic system. This has applications far beyond artistic works. Often sound and art is a very good excuse to spend time learning things that can be used for "real" applications later. For instance, would you prefer students build a clarinet playing robot or a robot that throws beer? There are plenty of examples of "useless" projects undertaken by undergrads -- but they then move on to produce useful results later in life.
Secondly, there is the psychological / human-centric part of the work: building robotics to mimic human gesture and human expression teaches us a lot about how we work.
This is called "basic" research: it doesn't necessarily consist of making something that DOES something, but it allows us to learn more about ourselves. Learning what techniques are needed for expressive clarinet playing implies that we know what those techniques are-- it implies we know what "expression" is, and that we understand much of the physics behind airflow and reed action.
This is interesting stuff. You say it is a clarinet that is "not being played very well." Well, WHY is it not being played very well? What can be improved in the playing technique? Why can humans do it better? Is it the lack of "expression" in playing (cognitive), or is something wrong with the airflow-reed interaction (physics). Or both?
This is physics and engineering and psychology all rolled into one amusing project. How can you say it's useless?
Robotic pianos have been around for over 100 years & they've never sounded as good as a human. After all this time they finally moved on to other instruments.
A colleague told me of this discussion and suggested that I give a brief explanation of the motivation for this project. I'm from the Music Acoustics Group at UNSW. We maintain a large web site http://www.phys.unsw.edu.au/music/ for the benefit of musicians, students and interested others. It has more details on the robot. The introduction on our site is aimed at a good high school student, but if you go deep enough it leads to our technical research papers. - Most of the time, we study real musical instruments, real musicians, the voice and the ear. Some of this is sponsored by companies (instrument makers, a medical device company, a museum), but much of it is curiosity research. - For us, the robot project complements one of our areas in which we study real musicians and how they play. We want to know, in some detail, why a real musician plays better and makes a better sound than a beginner. (Curiosity research, but with an obvious application in music teaching and sometimes instrument design.) - The robot is a tool for testing our understanding of the clarinet-player system. The current version is very primitive: it was put together in a hurry for the competition. But in the next year or so we shall use it to understand a range of questions: * Why does a clarinet reed squeak? How can you stop it? * What are the important parameters in a good sound? * How important are tongue position, soft palate, glottis? What are the best combinations? * How important is lip damping, and how does it depend on the reed? * What are the important parameters in fine pitch control? * What are the important parameters in expressive performance? * What is necessary to convey warmth? * What is necessary to follow a conductor? - To some of these, of course, we already have answers from our previous research. But we want to have more confidence in those answers. - So for the Music Acoustics Lab, this robot is a very useful tool. It was also a good project for two undergraduate students (Paul and Jean) in physics: a project that required a range of experimental and analytical techniques. The other groups in the robot team have different motivations. - For Mechanical Engineering, this robot was an interesting challenge. It was a good undergraduate student project for Kim: a range of questions to answer and difficulties to overcome. - It was also an interesting challenge for Mark, a Computer Engineering student Mark. In fact all of the students involved were highly motivated, worked well, learned a lot -- and had a good time. For university staff, this alone would justify the project. - For NICTA (a national research centre in ICT), the contest was a formal challenge. A good way of displaying expertise and applications in embedded systems, and a good way of inspiring students. (John Judge is from NICTA). - The team details and some more discussion is at http://www.phys.unsw.edu.au/jw/clarinetrobot.html Music Acoustics.
but can it play Crysis?
Most of the time, we study real musical instruments, real musicians, the voice and the ear. Some of this is sponsored by companies (instrument makers, a medical device company, a museum), but much of it is curiosity research.
For us, the robot project complements one of our areas in which we study real musicians and how they play. We want to know, in some detail, why a real musician plays better and makes a better sound than a beginner. (Curiosity research, but with an obvious application in music teaching and sometimes instrument design.)
The robot is a tool for testing our understanding of the clarinet-player system. The current version is very primitive: it was put together in a hurry for the competition. But in the next year or so we shall use it to understand a range of questions:
* Why does a clarinet reed squeak? How can you stop it?
* What are the important parameters in a good sound?
* How important are tongue position, soft palate, glottis? What are the best combinations?
* How important is lip damping, and how does it depend on the reed?
* What are the important parameters in fine pitch control?
* What are the important parameters in expressive performance?
* What is necessary to convey warmth?
* What is necessary to follow a conductor?
To some of these, of course, we already have answers from our previous research. But we want to have more confidence in those answers.
So for the Music Acoustics Lab, this robot is a very useful tool. It was also a good project for two undergraduate students (Paul and Jean) in physics: a project that required a range of experimental and analytical techniques. The other groups in the robot team have different motivations.
For Mechanical Engineering, this robot was an interesting challenge. It was a good undergraduate student project for Kim: a range of questions to answer and difficulties to overcome. - It was also an interesting challenge for Mark, a Computer Engineering student Mark. In fact all of the students involved were highly motivated, worked well, learned a lot -- and had a good time. For university staff, this alone would justify the project.
For NICTA (a national research centre in ICT), the contest was a formal challenge. A good way of displaying expertise and applications in embedded systems, and a good way of inspiring students. (John Judge is from NICTA).
The team details and some more discussion is at Music Acoustics.
My Palm pilot has the same piece of music as a sound for the alarm clock,
(the flight of the bumblebee) and plays it just as crappily.
But at this stage I think that was not really important yet, they'll get to the 'once more, with feeling' stage later.
I've blown on saxophones and clarinets just to try and make some noise. To have a machine getting this far is actually quite an accomplishment I think.
Now they just need to work on timing & emphasis....
The Bigger The Headache The Bigger the Pill
my favorite.... the robotic drummer: http://www.graffagnino.net/wwwpeart/ well... i guess I am a little biased :)
Since the present discussion includes issues about what is good/bad/possible/etc., this leads to the question of whether there is a way to *model* music - to shed light on issues like these, in a systematic way.
Our OMSModel project has been a very successful way to model music - individual works, and individual performances. OMS doesn't look at notes, rather it focuses on how music stimulates or affects a listener. It provides detailed analysis; it can provide numeric ratings (the lowest/highest rankings we have discovered so far are 3/680; the robotic clarinet performance is about a 30). It often provides very good "answers" to the big questions about music (e.g. "What distinguishes good music from bad music?"; "How can music be analyzed in a way that allows for variation in individual responses and preferences?"; "Does music convey meaning?", etc.).
OMS uses ideas from Marvin Minksy's "Society of Mind" (Music is modeled as organized multi-stimulation of a human's "musical agents"), and also some time-honored ideas from academic philosophy. To build out the OMS model (to come up with a good list of "musical agents"), we did in-depth analysis of some great musical works - starting with a couple of Beethoven symphonies and working out from there. Our analysis eventually included a broad range of music: Popular, traditional, world, jazz, classical, avant-garde, and more. We now have a good list of musical agents (somewhat over 100, and fairly stable). It turns out that Beethoven was able to provide over half of the agents; it may sound crazy, but LVB seems to have anticipated some of the ideas in "Society of Mind" (!!)
Detailed materials about OMS are are on www.OMSModel.com
Here are a few things that OMS says about the robotic clarinet performance: The performance rates at about a 30, which ranks it as better than Barney singing his song "I Love You" [18], and ranks it about equal with "It's a Small World" as experienced at Disneyland [31]. It is not up to the level of "The Chimpmunk Song" (Alvin and the Chipmunks, [43]). It is far short of "Baby Beluga" (Raffi, [100]), or a fun pop tune ("Rockin Robin" [120]). These numeric ratings are a rough estimate of the intensity and breadth of the stimulation of the music. Most mainstream pop music is in the range of 100 - 180. Anything 200 and higher is really intense and broad (e.g. the best performances by James Brown or by Miles Davis). 300 and up gets into the range of "mind-blowing" or "unforgettable". Over 300 can be life-changing!
A rating of 30 is a big accomplishment. A large amount of musical and technical analysis would be required to get to this level. To compare: If a child were to pick up a clarinet for the first time and try to play something, this would be about a 3. So, among other things, the Australia team has effectively simulated the result of a lot of clarinet lessons.
The effort to get from 30 up to 120 (= fairly good pop music, good performance) is roughly linear. I.e. if we were to assess the total effort to accomplish the robotic clarinet performance (including prior efforts by other researchers), then multiply by 4 to get the level of effort to get to fairly good pop music. (We haven't estimated the effort to get above 120, but my guess is that the effort starts to become steeply logarithmic at some point shortly after 120. So probably a long long way to get to the level of Richard Stoltzman or Benny Goodman.)
Just imagine if all this effort had been spent on one of the millions and millions of things that a human cannot already do better. We could have had ALL the same "because I can" and STILL YET yielded some usefulness in the end.
Everyone seems to simply go "cool! robots! OMGoneoneone" and misses the opportunity cost entirely.