Sometimes this Everyday Musician blog will offer some glimpses into the future of technology for musicians. This one came across my desk today. Researchers (including Mark Bocko) at the University of Rochester have simulated a musician playing an instrument.
The claim of simulating a musician might lend itself to controversy, as well as humor along the lines of drummer jokes. I’ll let you all have the fun of making up jokes about this your blog comments. This is very interesting stuff, that I’ll consider here seriously.
For many years, there has been a method of simulating instruments that is termed “physical modeling.” The physical modeling of a wind instrument, such as a clarinet, is in part a computerized simulation of the wind tunnel in the clarinet, with wave pressures of air bouncing around. The physical modeling of a violin string is a simulation of the string vibrations that are picked up by the body of the violin as the bow crosses the string. Physical modeling has been available in commercially available electronic (MIDI) keyboards as early as the mid-90s. When you hit a key soft or hard on the keyboard, your action is interpreted as blowing soft or hard on the clarinet, or bowing soft of hard on the violin string.
So, what’s new here? Two days ago, the researchers announced at the International Conference on Acoustics Speech and Signal Processing that their computer simulations not only mimic what’s happening inside the musical instrument but also mimic the physical actions of the musician himself playing that instrument.
For example, this University of Rochester project simulated the physical activity of musician playing the clarinet, including “the fingerings, the force of breath, and the pressure of the player’s lips to determine how they would affect the response of the virtual clarinet.”
This simulation of the musician playing the instrument offers a far richer way of controlling the virtual instrument. Instead of controlling the sound of the clarinet by just hitting keys soft or hard on the keyboard, and perhaps using a joystick on the electronic keyboard, the virtual clarinetist actually has a virtual mouth cavity, and lung blowing air, and tongue articulating the notes.
As a keyboard player, I’ll be interested in how this technology might evolve to enable a keyboard player to “pretend” he is a clarinetist, by controlling the virtual musician portion of this simulated musician-instrument combination.
15 responses so far ↓
1 Mark Bocko // Apr 4, 2008 at 7:06 am
I’m glad to see your interest in our work. I think that you have identified the most interesting part of this research - what might future human interfaces to such models look like? First, one aspect of our work that is a bit lost in the details is that our physical model is experimentally derived from data from a real instrument measured in the lab. This will enable the computer model in the synthesis engine to be a detailed representation of a known fine instrument. The second thing is that for existing physical models it’s difficult to find the set of parameters that make good sounds, it takes a lot of tweaking. The third thing is that we use our model as an analysis tool to figure out what the musician must have done, how hard they blew into the instrument, the embouchure force, etc..at every instant to create a musical line. We record the time-series of the control parameters as a way of encoding the sound. This brings us back to the first point, is it possible to identify a library of musical gestures that could be available to the musician, through a keyboard or other interface, to enable them to create musically pleasing lines? We’re currently thinking about this and other ways (like a Wii interface) to enable a musician to control our types of models. We’d love to hear any ideas out there. Thanks for your interest.
2 Everyday Mark // Apr 4, 2008 at 12:01 pm
Hello Mark Bocko,
Thanks for replying to this blog post about your fascinating research in the area of computer simulation of both the physical musical instrument and the performing musician. It is rare opportunity for everyday musicians to get a chance to meet one of the researchers advancing computing technology in the area of music.
It would be pure hubris on my part to respond, after only brief consideration, to one of the important questions your team has spent a lot of smart time researching: “Is it possible to identify a library of musical gestures that should be available to the musician, through a keyboard or other interface, to enable them to create musically pleasing lines?” Therefore, I’ll respond.
From an everyday musician’s point of view.
Everyday musicians play at least one kind of instrument, and/or sing. The most common instruments are, of course, the guitar, piano, and voice. One might think of a musician’s first instrument, or voice, as his native language.
What musicians would love to do is be able to speak in other languages, i.e., play other instruments. A piano/keyboard player would love to be able to expressively play a violin, guitar, or saxophone. A violin player would love to be able to expressively play a saxophone (which I first enjoyed hearing ten years ago by a musician playing a MIDI violin.)
What musicians might hope for is a Star Trek-like universal language translator, where one can go to any galaxy and communicate perfectly in another language by speaking one’s own native language. The Star Trek universal language translator picks apart the sound waves of the speaker; probably then translates that into an intermediate universal language (of “musical gestures”) that only the computer understands; and then spits out a translation into the other language.
This universal language translation approach seems to be within the realm of current possibility as an approach that would allow a musician playing his native instrument, or singing, to virtually play another instrument. Although no researchers have done a good job picking apart sound waves (polyphonic) of musical ensembles, there has been fairly good progress made with analyzing sound waves (monophonic) of solo instruments and voice. The task at hand here would be to translate incoming sound waves from a variety of alternative instruments, or voice, into the intermediate language of “musical gestures”. It sounds like your team is already well on its way towards taking it the rest of the way, that is, to apply the generalized musical gestures to specific virtual instruments that you have modeled.
An alternative approach would be to offer musicians a new, easily playable universal instrument which directly speaks the language of “musical gestures”, such as you suggested with the Wii interface. This approach would be particularly appealing to new musicians, who are not already highly committed to their native instrument or voice.
In investigating this approach of a new universal instrument, I can highly recommend that your team take a look at the Thummer instrument invented by James Plamondon. See http://www.thummer.com. The Thummer instrument receives musical gesture input from the musician not only via joystick but also gravitational motion of the entire instrument as the musician waves the small accordion-like instrument through space. Also, the Thummer offers an elegant keyboard layout that is completely neutral with respect to harmony. With the Thummer, the finger pattern for a C#-major chord is exactly the same as for a C-major chord; and the finger pattern for playing a C# major scale is exactly the same as for a C-major scale.
I hope that this small meeting room in the huge conference world of the Internet might serve as a good place for you to meet some everyday musicians who, like me, will be highly enthusiastic to see your virtual instrument/musician technology find its way into their hands in the marketplace.
Cheers
– Mark
3 Djim Tio // Apr 4, 2008 at 4:29 pm
Hi Mark
Thanks for introducing me ( and others I hope ) into the latest remarkable developments in sound engeneering.I am quite convinced that soon it will be possible to emulate a variety of instruments just by “playing ” your PC keyboard,which is , according to me ,a frightening
idea.
What I mean to say is, catching the physical model of an instrument does not necessarily include catching the soul, interpretation , the technique,the feelings etc. of the musician knowing the possibilities and impossibilities of his instrument.
As an example, I and every instrument player, can hear and see ( in Notation programs )
that certain instrument parts sound beautifully
but impossible to reproduce or play practically.
I am quite convinced that Cpt.Kirk can handle the “Enterprise ” in whatever situation,but commanding the ” Milennium Falcon ” is another cook.
Just joking,of course,but I really like your approach of opening one´s mind to think about
the future of music and musicians.
Thanks and regards
Djim
4 Jim Plamondon // Apr 5, 2008 at 1:46 am
Gentlepersons,
I am honored to have the Thummer mentioned in such illustrious company.
One of the Thummer’s design goals was to provide a highly-expressive polyphonic interface to electronic music synthesis, especially physical modeling (with waveguide synthesis). Its other main design goal was to be easy to learn, so that its expressive power could be as accessible to as many people as possible.
Mark, I would be happy to send you a prototype if I had one to spare, but I do not — they’re all loaned out at the moment. However, you’ll be happy to know that they are in the hands of people who are looking into this very issue. Stanford’s Julius O. Smith, father of waveguide synthesis, has one, for example, as does the Electronic Music Foundation’s Joel Chadabe (neither of whom should be presumed from the above to have endorsed the Thummer at this time; they’re just checking it out).
The best I can do right now is to point you at Thumtronics’ website, http://www.thummer.com. I apologize for my tardiness in bringing the Thummer to market.
Could you direct me to a version of your recent paper on the Web? I’m particularly interested in understanding the extent to which your model binds synthesis variables together into musically-pleasing multivariate spaces. That could be very useful for novice musicians. It might also help stimulate artistic creativity by clearly defining the “rules to be broken.”
I am also intrigued by your notion of creating “a library of musical gestures that could be available to the musician, through a keyboard or other interface, to enable them to create musically pleasing lines.” This reminds me of the issue of having a given musical gesture be interpreted consistently:
1) by different synthesizers of the same acoustic instrument (e.g., the clarinet), and
2) by synthesizers of different acoustic instruments (e.g., clarinet and trumpet).
Having a single set of expressive control gestures that is invariant across a wide (but bounded) range of synthesizers, let alone across synthesis algorithms, is, so far as I know, an unsolved problem. I’d welcome your thoughts on it.
Thanks for the interesting research — and the friendly discussion!
Jim Plamondon
CEO, Thumtronics Inc
The New Shape of Music(tm)
http://www.thummer.com
Austin, Texas
5 Everyday Mark // Apr 7, 2008 at 10:37 pm
For those who might not be familiar with some of the terms above, here’s a quick summary:
The Thummer is sort of a hand-held version of an electronic (MIDI) keyboard, with a keyboard arrangement that looks similar to that of a little accordion. The note layout keyboard is designed to be optimally easy to learn to play. The Thummer also lets you control the quality of the sound with a thumb control (hence the name Thummer) as well as motion of the entire instrument up/down/right/left.
This Thummer instrument can serve the same purpose as a MIDI keyboard to let a musician simulate the playing of other “virtual” instruments– strings, woodwinds, brass, percussion.
The above discussion suggests a possible match between Jim Plamondon’s Thummer instrument and the virtual instrument technology that Mark Bocko is working on at the University of Rochester.
I wish this stuff were all available in the marketplace today.
Cheers
– Mark
6 Bernie // Apr 8, 2008 at 5:15 am
I am happy to see these innovations, and am sure many of the existing problems inherent in this degree of realism will be overcome.
Having said that, someone must play the devil’s advocate. As we all know, many synth manufacturers have introduced everything from wind controllers, modelling programs, to guitar mode. The biggest problem seems to be that everyday musicians can’t obtain the true benefits of even these programs without having knowledge of how the instrument is actually played.
Won’t this be magnified with such advanced programs ?
7 MMA // May 8, 2008 at 5:47 pm
Hello everyone.
I can’t help but respond to a request for “a library of musical gestures that could be available to the musician, through a keyboard or other interface”.
MIDI is a technology that uses the gestures a musician makes while performing on music keyboards. The various gestures that MIDI supports are “events”, each represented by a string of numbers. An instrument that receives a MIDI event message is responsible for performing the action that corresponds to that gesture… for example the gesture “Pressing the Middle C Key with maximum velocity” will tyically produce some sound pitched at middle C). The definition of the sound (is it a guitar sound, piano sound, or bird call?) is determined by a separate “sound selection” gesture. In this way, MIDI allows performances to be fully editable (as well as scorable) and creates a compact represenation compared to an actual recording. Plus with careful rendering, a MIDI performance can be the equivalent of an audio recording.
The technology was developed by makers of music keyboards,so it is very well suited for that market. Unfortunately, many people who play other kinds of instruments (guitars, strings, winds) find that MIDI’s “Key-on/Key-Off” model for sound generation does not always allow for direct translation of all the different means of expression in those instruments. Still, as one writer said above, there are MIDI-equipped guitars and wind instruments none-the-less. Over it’s 25 year history, many innovations have been made in realistic interpretation (rendering) of MIDI, and many new event mesages have been defined, but MIDI is basically the same as it was.
Now the MIDI industry is looking at how to better represent more kinds of instruments… for example, if new messages could be defined that better represent the gestures of a violinist. The challenge is not just figuring out what gestures are appropriate to each instrument, but also how to quantify them (define their attributes and range), and how to harmonize them so that each type of playing technique will also work on other instruments.
From a technology perspective, what’s needed is a message format that is of sufficient depth (for accuracy) and breadth (for scope ) to be able to represent a significant body of instruments and playing styles. Then there must be a dictionary that describes the messages for each of the ways that an instrument can be affected through human intervention. We would call this a “dictionary” since it would merely list and describe gestures, as opposed to a “library” which as commonly used in the technology industry refers to software that accomplishes some specific task.
The benefit of the dictionary vs. the library is that the dictionary would allow products to interoperate regardless of what library they were using, which protects consumer investment and allows manufacturers to innovate. We understand that some people may actually want the library because they don’t have any way to render the gestures, but I wanted to point out that the dictionary is more important for building a market.
Tom White
MIDI Manufacturers Association (MMA)
8 Casey Zalusky // Jul 9, 2008 at 5:39 pm
Take a look at this automatic violin (http://www.youtube.com/watch?v=zk3QBCo5dUQ). It’s very cool!
9 Everyday Mark // Jul 11, 2008 at 10:14 am
Hello Casey,
This is indeed very cool! A violin playing itself. Non-mechanically!!!
Upon brief Googling, I wasn’t able to find the company Forever Music R/D (’love that company name) that is developing this violin playing mechanism.
My guess is that it works electromagnetically by stimulating vibrations in the strings. It’s a very obvious way to auto-play a stringed instrument… once some clever person has shown the rest of us the obvious idea
Cheers
– Mark
10 hxflyer // Dec 21, 2008 at 9:31 pm
Hello Mark Bocko, Mark, Djim, Jim Plamondon , Tom White
First I have to say you all awesome, I search the simulation of instrument from google and get this website, I am a computer graphic programmer, doing a research on how to use digital gesture capture device to get musician’s input and use computer to simulate the sound out,I did few months parttime research,and this is my abstract ideas:
1.instruments is a kinds of human interface for muscian to input data through physical movement which generated from human.
2.instrument receive the data from muscian and output as sound wave through physical transformation.
3.the input mechanism of all the instruments are limited by the sound wave physical transformation of the instument it self,for example,you can not use the way of input data into piano to get output as sound wave of violin.
4.our goal is to find out a common digital encode mechanism which be able to describe the input data and then simulate the sound wave transformation in computer, and finally produce the sound as result.
As Tom White said, the MIDI is not enough for record all the data of muscian’s input, and as Mark Bocko said the simulation of physical tranformation also have long way to go. I have spend 3 month to read some of the thesis produced from MIT and NASA, I have to say they did pretty good job and we are more and more close now, to simulate the sound of instruments and to record the muscian’s input is not far away from realistic.
what we are tring to do is not try to simulate muscian’s playing,because there is noway to simulate the emotions and human intelligence of creat art. what we tring to do is to extend the muscian’s potential and eliminate the limitation of instruments ,I see a lot of people in this world have great feeling and ideas of music but their body is not facility enough to play the music and I also see to be a great player need spend decades to practice, the reason exist is because the interface is not nice enough, get used to a kinds of instrument is a very long way. and also the different interface of different instrument are limited by the sound transform hardwares of itself.
If we can find a way to solove the data input and transform simulation, then you can use piano to play violin and use violin to play piano,or make the sound half piano and half violin, or even use some new interface( keybord? mouse? touchscreen? scanner?camera? )as input to make music,just need to define a set of commands. I believe when that day coming the music would be more magic and fascinating, and people can use the best way of making music whatever they like.
I have to say piano is invented 300 years ago and violin is invented 500 years ago, even people improved the technology of the hardware inside the instruments, we haven’t change the interface and tamber too much, find a easy way for human to creat music and make the special tamber no instument in real world beable to make ,it must be a revolution, isn’t it?
sorry my english is not good enough.
Huang Xiang
researcher from PeKing University Digital Art Department , china
I am looking for share more ideas with people doing the same research or muscian have different thinking. mail: hxflyer#gmail.com
11 Everyday Mark // Dec 23, 2008 at 4:34 pm
Hello Huang Xiang,
> I am … doing research on how to use digital
> gesture capture device to get musician’s input
> and use computer to simulate the sound out
This is interesting research that you are working on.
> If we can find a way to solove the data input
> and transform simulation, then you can use piano
> to play violin
As much as I love the piano (that’s the instrument I play), I must admit that the violin is far more expressive than the piano. There’s more musical information in an expressive violin performance that there is in an expressive piano performance. You can pretty much fully represent an expressive piano performance with no more information than the timing of which keys are hit on the piano at what velocities, plus pedal data. I can capture a MIDI performance on my Yamaha Disklavier, and play it back with very little loss of musical information (expressiveness). It is not possible to encode all of the musical expression of a violin performance with just MIDI pitches, timing, and velocities. That’s why we have MIDI patches and controllers for different types of string strokes (degree of vibrato, staccato, detache, harmonics, col legno, etc.)
Therefore, I’m quite skeptical that there is a good way to use the piano (MIDI keyboard) as a an ideal expressive input device to simulate the violin. Going the other direction would work, using the violin as an input device to generate piano sounds; but that’s not an interesting problem to solve.
Still (continuing with the example), it is a delight to a keyboard player to be able to use the keyboard as an input device for producing violin sounds. The keyboard player must accept, however, that his instrument won’t be able to generate all of the nuances of the expressiveness of a violin.
Generalizing the above, different types of instruments can probably be arranged in a spectrum of degreess of expressiveness. But this would not be a one-dimensional spectrum with probably the human voice at the top end of expressiveness. Rather, it would be a multi-dimensional spectrum, in which different instruments excel in different ways such as:
* how much the harmonics change according to the initial attack
* how much the sound wave envelop evolves over time throughout the entire duration of the note
* how much control the musician has over the sound after the initial attack. (For example, the acoustic pianist only has a few options at his disposal, such as fluttering the pedal or holding down keys to pick up sympathetic vibrations.)
With this generalization in mind, perhaps a useful restatement of the challenge is this: How should a music input device be designed that is easy to learn to use, and yet which also supports a high degree of expressiveness?
I think the most challenging factor here is simply pitch selection. There is simply no easy short-cut to picking out pitches from the western 12-tone system in real-time. The keyboard pitch layout of the James Plamondon’s Thummer instrument seems to me to be the optimal pitch input device. I’m disappointed that James hasn’t had success so far in promoting this new instrument.
If the musician is willing to give up precise control of pitches, then a huge world opens up. For example, consider turntable scratching. The turntablist can control what specific pitches are played, but he has lots of control over rhythm and musical structure of the performance. Perhaps a good general term for all such musical performances devices would be “conductor’s device”.
Huang, have you considered musical input devices that enable individuals who don’t play any instrument to still expressively “conduct” music?
What happened to those electronic conductor’s batons I saw at NAMM several years ago?
Cheers
– Mark
12 hxflyer // Dec 25, 2008 at 6:27 pm
Hi Mark:
Thank you so much for give me these great thinking, I don’t play any instrument , I have admit that I’m sort of blind in musician’s way of thinking，Here is some idea about our thinking:
about the the huge different between the instruments you mentioned, it was true that input for violin is far more complicate than piano, piano keyboard can not record all the violin input data, this is the reason that we need capture more information at keyboard than the old way, the way of doing this, is decompose the violin input date into a group of individual metadata, and convert the metadata into a group of keyboard dictate as input.(this url is a researcher in MIT doing violin simulation and data recording : http://web.media.mit.edu/~young/publications.html
) Consider there are too much data, as you can see, we can not use MIDI data structure ,(the MIDI is record data by events, but we need record the data by sample) ,and also need use several different dimensions to record keyboard input, the sensor is not only record pressure and velocity, but also we must define some group key and hotkey gesture to recuperate the missing data. In my opinion , input for violin and input for piano are all gesture movement from human, change one gesture movement to another is like change one language to anther, It’s definitely need re-organize the data of represent and somehow would loss some data. Be honest I have no idea what the result is, maybe it wouldn’t be able to make perfect transfer for using the technology what we have so far, but we still be able to make it sort of “work” for games and entertainments now. That is the reason why we want implement it on wii, maybe the recent product would be like toy for the professional musician like you, but in future ,I believe the perfect simulation is possible.
As you said the another problem is the mental emotion and expression would change if the player change the gestures, the feeling of play violin is different as play piano, Does the musician like to change the feeling? or does the musician still be able to play good music when the feeling changed? I have to say I have no idea either. But this is worth to try.
I have talked with a violin player, he said when he put violin on the shoulder, his ear very close to the violin, and he’s shoulder and finger are very sensitive to catch the minor movement and vibrations, these kind’s of feedback from instrument help him control the instrument, if the feedback gone, does he still be able to control? And how to simulate the feedback from instrument with totally different interfaces? This is the new question you bring to me.
The another problem of how to cooperate with singer and instrumental ensemble, we haven’t think that far, the technology is still in the first step of record input data and simulate voice of instrument. I have no answer for that.
“How should a music input device be designed that is easy to learn to use, and yet which also supports a high degree of expressiveness?” This is the original idea at the beginning of research ,but I found out that to achieve this goal we have to separate the input interface and sound output hardware first, the way of separate the two steps ,is to use computer to simulate.
About enable normal individual to play good music, I don’t think it is really possible, No matter what kinds of interface is, the player need to input huge amount of data every second, and that require player must master the interface anyhow. And also I think master the interface doesn’t mean be able to play good music, player still need talent and feeling on music. But my idea is if we can make the interface easier and make the musician focus on music and sound itself not the skill of input data into instrument would make music different. And if we can separate the interface and the sound generate hardware, people can choose the interface themselves.
Be honest I don’t think our research result would good enough for simulate real music, at least we can not achieve that goal in these several years, A lot of research project get negative conclusion, but the important thing is we try it,as technology improving everyday, I believe in future this is possible.
Merry Christmas!
Huang
13 Everyday Mark // Dec 28, 2008 at 12:03 pm
Hello Huang,
Have you looked at James Plamondon’s Thummer instrument? It is demonstrated at http://thumtronics.com/
I believe the Thummer offers an excellent balance between expressiveness and ease-of-playing. Thummer music keyboard layout is well-designed for selecting pitches. The Thummer user controls expressiveness with: (1) a thumb control similar to a joystick on a MIDI controller, and (2) movement of the instrument through 3D space, similar to how the Nintendo Wii operates.
Cheers
– Mark
14 James Plamondon // Dec 28, 2008 at 2:57 pm
Dear Huang, et al.,
You wrote that your original reseach problem was stated as ““How should a music input device be designed that is easy to learn to use, and yet which also supports a high degree of expressiveness?”
The answer to this question is “the Thummer” (especially when combined with the ThumMusic System of notation). If you want to make a major leap forward, you have to look beyond past instruments & notation, because they are a big part of the problem.
Ease of Learning & Use
1. The Thummer’s isomorphic note layout has the same fingering in every octave, key and tuning (of the syntonic temperament). Combined with tonic solfa and electronic transposition, isomorphism (a) reduces the set of symbols and gestures that a novice musician must learn, and (b) relates these symbols and gestures to each other in a simple, visual, tactile, and logical manner, thereby “engaging more brain to learn les stuff.”
2. The Thummer’s controls can be progressively enabled, enabling a novice to learn one hand’s keyboard first, then the other, then a single thumb-stick, then the other, then a motion-sensing axis, then another, and so on. That is, the Thummer’s complexity can grow with the student’s skill. This is quite different from (say) a saxophone or violin, the full complexity of which a novice cannot avoid, even on her first note.
3. The Thummer leverages user-interface skills that non-musicians already have. The keyboard’s buttons are spaced to match the keys on a computer keyboard; the thumb-operated joyticks are familiar to anyone who plays video games; controlling the movement of one’s hands through space is an inherent human skill. The Thummer applies these existing user interface skills to music-making, rather than requiring musical novices to acquire entirely new gestural skills.
Expressive Potential
The expressive potential of any music-control interface can be approximated by the number of independent continuous variables that its performer can simultaneously control. Most instruments offer few such variables — three, four, or five, often plus a couple of discrete variables (such as foot pedals being on or off). The Thummer offers more “expressive potential” than any other instrument: up to a dozen such independent continuous variables. And, it’s polyphonic, like a piano.
International Potential
The Thummer’s isomorphic keyboard has the same fingering in every tuning of the syntonic temperament, which includes the modern Western standard 12-tone equal temperament, but also many other interesting tunings such as indigenous Thai, Indonesian, and African tunings. Ancient Chinese bronze bells were tuned in 1/4-comma meantone, which is a syntonic tuning. Whether the tuning has 5, 7, 12, 17, 19, 31, or even more notes per octave, it still has the same fingering on the Thummer keyboard. No other instrument offers this, plus expressive power, too.
Creative Potential
Because the Thummer has the same fingering in every tuning, its players can change the instrument’s tuning during performance, sliding smoothly from one tuning to another as an expressive effect. This has not previously been possible, and enables many novel musical possibilities.
Commercial Potential
The Thummer is patented in the USA, but not in China, so Chinese researchers and manufacturers can make, sell, and use Thummers all they want, outside of the USA. I would be happy to help your commercial partners sell Thummers in the USA, too. The current economic downturn is the ideal time to invest in developing a new technology such as this, which can lower the Chinese government’s cost of music education and give its musical instrument industry the opportunity to out-innovate its non-Chinese competitors.
Huang, you have a great opportunity to make a great deal of money, to advance the state of the art, and to make the world a better place, by putting together a deal between Peking University (music department and engineering department) and a local manufacturer of electronic musical instruments to make and sell Thummers. Your electrical engineering students can design the Thummers; your music education students can design the courseware; your software engineering students can write the code, and the local manufacturer can put it into production. You put the deal together, and I’ll send you some prototypes and design documents.
Interested?
Thanks!
Jim Plamondon
CEO, Thumtronics Inc.
Austin, Texas
15 Jessica Wang // Jul 14, 2009 at 2:04 am
Dear Sir,
How are you?
We are a professional enterprise to produce various musical instruments including string instruments and wind instruments for many years in China. Our main products are violin,white violin,color violin,electric violin,viola,cello,double bass, guitar, wood bows,carbon bows,bow parts,brass instruments and all kinds of tools and accessories, especially our bow hairs are very popular in the bow fittings, for details,please visit our websites: http://www.katamusic.com.cn (http://www.katamusicwindinstrument.com).
The high quality materials and our experts’ unique inspiration and rich experience make the products not only graceful but also very reasonable. Based on expanding market,our price is very competive, our products will say themselves.If any article is interesting to you,please feel free to contact us .
Looking forward to hearing from you.
Yours truly,
Jessica
Kata Music.Co.Ltd
Tel:0086-318-7877110
Fax:0086-318-5181851
Web: http://www.katamusic.com.cn
Email: katamusic.accessories@yahoo.com
Msn: wangjianhua71@hotmail.com
Leave a Comment