Each and every sound we hear in this world has its own uniqueness, but we don’t often use all of these sounds to create music, because it’s pretty much hard to capture these sounds with their true essence and, Of course, the music of our generation is not just about one particular sound; the music industry is in an unceasing quest for new sounds that could help them create the out-of-the-world experience we expect of them. So what? Now that technology has taken the reins of our world, almost everything is being re-imagined with technology, and then why not think of a new way to create music through machine learning as well? The answer is NSynth Super.
What is NSynth Super?
NSynth Super is an open source experimental instrument which utilizes machine learning to analyze music and come up with unique different sounds. It’s a part of Magenta, a Google’s research project on using machine learning in the process of creating music and art. Besides,Magenta built this particular instrument completely on TensorFlow.
Several years ago, we were introduced to Hartmann Neuron—A Neuronal Re-synthesizing Keyboard which used Artificial Neural network, which was also used to re-synthesize a new sound, with quite a bit of limitations on its side, which resulted in the termination of its manufacture in 2005. Then why create NSynth Super? NSynth Super functions as a physical interface for the NSynth algorithm which was released back in 2017— an algorithm which can learn the unique acoustic quality of each sound and can combine sounds to create a completely new sound with the essence of the sounds involved. To be exact, the algorithm doesn’t just mix the sounds together, but blends the acoustic qualities of sounds to create a fresh new sound. Using deep neural networks, NSynth Super possesses the very capability to create a new sound that could be played back with several timbres, pitches and tones. This instrument’s preciseness and performance exceeds everything, seeing that it has been trained over 300,000 instrument sounds, which is so far the largest data set of musical notes that is publicly available.
So, it creates new sounds, Is that all? Actually, no. The big news is that all the hardware specification, interface, design template and source code related to NSynth Super have been made open source, which is now available on GitHub. This has opened the door to anyone who wishes to experiment with music to make their own NSynth Super by themselves, since it’s not yet available for purchasing.
How to use NSynth Super?
Using NSynth Super doesn’t actually need a manual, although it is filled with a lot of pretty knobs and dials. First of all, you have to upload a pack of 16 prepossessed source sounds and simply dragging your finger across the touch screen will create new acoustics. Furthermore, you could use the dials to refine the sound space indicated on the screen. Also, it is compatible with all MIDI sources.
Each and every technology invented so far, was aimed to assist, to enhance and to work together along with human. Even though the present day technology is surpassing the human abilities in art and music, we should always keep in mind, that technology should always assist the artists, not replace them. Let’s hope it never comes to that…
To learn more about NSynth Super: