Anyone who has through the process of learning to play a musical instrument for the first time, or listening to someone attempting to do so will know that it can be a rather painful and frustrating experience. [Alessandro Perini] apparently couldn’t get enough of the sound of a first-time musician, so he created a robot to play the melodica badly for hours on end, as demonstrated in the video after the break.

The project is appropriately named “AI’ve just started to learn to play”, and attempts to copy every melody it hears in real-time. The robot consists of the cartridge carriage from an old printer, mounted on a wooden frame to hold the melodica. The original carriage used a DC motor with an encoder for accurate movement, but since position accuracy was not desirable, [Alessandro] ditched the encoder. Two small trolley wheels are mounted on the cartridge holder to push down on the melodica’s key. A bistable solenoid valve controls airflow to the melodica from an air compressor. The DC motor and solenoid valve is controlled by an Arduino via a pair of LM298 motor drivers.

A host computer running software written in Cycling ’74 MAX listens to the melody it’s trying to imitate, and send serial commands to the Arduino to move the carriage and open the solenoid to try and match the notes. Of course, it keeps hitting a series of wrong notes in the process. The Arduino code and build instructions have been published, but the main Max software is only described briefly. [Alessandro] demonstrated the robot at a local festival, where it played YouTube tutorial snippets and jammed with a local band for a full 24 hours. You have to respect that level of endurance.

If listening to less error-prone electronically controlled instruments is more to your taste, listen to this building-sized pipe organ play MIDI files.

At Arduino Day, I talked about a project I and my collaborators have been working on to bring machine learning to the maker community. Machine learning is a technique for teaching software to recognize patterns using data, e.g. for recognizing spam emails or recommending related products. Our ESP (Example-based Sensor Predictions) software recognizes patterns in real-time sensor data, like gestures made with an accelerometer or sounds recorded by a microphone. The machine learning algorithms that power this pattern recognition are specified in Arduino-like code, while the recording and tuning of example sensor data is done in an interactive graphical interface. We’re working on building up a library of code examples for different applications so that Arduino users can easily apply machine learning to a broad range of problems.

The project is a part of my research at the University of California, Berkeley and is being done in collaboration with Ben Zhang, Audrey Leung, and my advisor Björn Hartmann. We’re building on the Gesture Recognition Toolkit (GRT) and openFrameworks. The software is still rough (and Mac only for now) but we’d welcome your feedback. Installations instructions are on our GitHub project page. Please report issues on GitHub.

Our project is part of a broader wave of projects aimed at helping electronics hobbyists make more sophisticated use of sensors in their interactive projects. Also building on the GRT is ml-lib, a machine learning toolkit for Max and Pure Data. Another project in a similar vein is the Wekinator, which is featured in a free online course on machine learning for musicians and artists. Rebecca Fiebrink, the creator of Wekinator, recently participated in a panel on machine learning in the arts and taught a workshop (with Phoenix Perry) at Resonate ’16. For non-real time applications, many people use scikit-learn, a set of Python tools. There’s also a wide range of related research from the academic community, which we survey on our project wiki.

For a high-level overview, check out this visual introduction to machine learning. For a thorough introduction, there are courses on machine learning from coursera and from udacity, among others. If you’re interested in a more arts- and design-focused approach, check out alt-AI, happening in NYC next month.

If you’d like to start experimenting with machine learning and sensors, an excellent place to get started is the built-in accelerometer and gyroscope on the Arduino or Genuino 101. With our ESP system, you can use these sensors to detect gestures and incorporate them into your interactive projects!

Bespoke Electromechanical Instrument was built by Jay Harrison as part of a dissertation undertaken on the Creative Music Technology degree course at Staffordshire University. The instrument, running on Arduino Mega 2560 is designed  to allow each note to be independently placed in a space:

The project involved the creation of an electromechanical system capable of autonomously playing a bespoke Lithophone musical instrument. The underlying idea was to create a Lithophone that allowed the audience to literally step inside it, giving a unique spatial and acoustic surround experience. Designing an autonomous electromechanical was the thought to be the most effective and reliable solution to achieving this.

The Arduino Mega 2560 was used to interface Max/MSP with the physical circuitry. Control messages/signals would be sent out of a Max/MSP patch using Maxuino, these signals would then be interpreted by the standard firmata sketch loaded onto the board and would go on to trigger and control the 24 rotary solenoids and 24 servo motors that work to produce the notes.

In the video below the instrument is arranged in a 24-foot surround configuration and the audience is invited to experience the instrument from within offering a unique spatial dimension to the Lithophone intended to completely envelop the listener:

Rodrigo Carvalho is a designer and interactive new media artist from Porto (Portugal) researching on real-time relations between sound, image and movement in audiovisual interactive spaces. He submitted to Arduino blog his latest project “Into the void” running on Arduino Mega:

First version of “Into the Void”, a series of audiovisual installations exploring physical structures creating light and shadows with immersive audiovisual spaces.
A array of triangles is placed on the floor, each one has a LED strip on the back which are connected to an Arduino. On MAX, series of random numbers generators and different probabilities trigger a signal to each triangle, making it turn on or off.
At the same time that a triangle is triggered a MIDI message is sent to Ableton Live and plays notes on a MIDI Instrument and an OSC message is sent to Processing for the Visual output.

The installation was created in collaboration with Ana Duarte, André Sousa and Daniel Correia and you can find more info at this link.

[vtol] is quickly becoming our favorite technological artist. Just a few weeks ago he graced us with a Game Boy Camera gun, complete with the classic Game Boy printer. Now, he’s somehow managed to create even lower resolution images with a modified typewriter that produces ASCII art images.

As with everything dealing with typewriters, machine selection is key. [vtol] is using a Brother SX-4000 typewriter for this build, a neat little daisy wheel machine that’s somehow still being made today. The typewriter is controlled by an Arduino Mega that captures an image from a camera, converts it to ASCII art with Pure Data and MAX/MSP, then slowly (and loudly) prints it on a piece of paper one character at a time.

The ASCII art typewriter was recently shown at the 101 Festival where a number of people stood in front of a camera and slowly watched a portrait assemble itself out of individual characters. Check out the video of the exhibit below.

Cata Sopros is interactive sound installation running on Arduino Uno and created by Elas Duas, a multidisciplinary studio based in the city of Guimarães (Portugal). If you translate the title from portuguese it means: Breathe Catchers. In fact the project is a collective musical instrument made with paper windmills transforming the users’ breathe into sounds:

The windmills have inbuilt electret microphones that were connected to an Arduino Uno. The sensor data was then sent to MaxMSP and the sounds were played with Ableton Live. The video was shot at the cloister of the beautiful Alberto Sampaio museum in Guimarães, Portugal.

Enjoy the video:

Hive (2.0) is the second iteration of an interactive sound sculpture consisting of fifty speakers and seven audio channels. The sensors detect the proximity of people and Arduino manipulates audio according to it.

It was created by Hopkins Duffield, a Toronto-based collaborative duo exploring ways to combine both new and familiar mediums with artistically technological practices. In this work they used Arduino Uno together with Max 6 / Max For Live.
Check the video to listen to the sculpture:

Read the full article on MAKE

Read the full article on MAKE

Read the full article on MAKE