MACHINE LEARNING

19 May 2016

A link for Musik Tech Fest provides a number of interesting projects – http://musictechfest.net/ 

Toa Mata Band is known as the World’s first LEGO robotic band controlled by Arduino Uno, which is hooked up to a MIDI sequencer. In this video (the fourth episode) the robots are playing another tribute, this time to the coolest electronic music duo ever. Enjoy!
The musical gears and apps used in this episode are: Ableton Push\Live, Arduino Uno,Fender Precision Bass, Finger BassLine, Coron Rockaku-kun, Korg DS10, Nintendo DS,Boss HC-2,Moog Animoog. The performance was recorded live in July 2015.

Wintergatan – Marble Machine – http://www.wintergatan.net/#/news

18 May 2016

A peer (Ginger) shared a link to one of her works in an exhibiton where other interesting installations and artists are represented –  http://www.surfaces2016.com/

The Cymatics Theremin is an interactive audiovisual installation that explores the relationship between sound and the visual patterns created by sound vibrations. The Cymatics Theremin is played by changing the distance between your hand and the sensor, which smoothly controls the frequency and amplitude of the audio signal that is played. The sound creates vibrations that generate shifting surface waves within a dish of water. As a person plays the instrument, they can enjoy seeing the forms and patterns of various sound vibrations in real time on the water’s surface, or projected as a large-scale visual display – http://www.synthestruct.com/hydromancy/ 

 Computer Talk –  is a creative collaboration between Reeps One (audio) and Synthestruct (visuals). The sound was composed by Reeps One using only his voice as an instrument (aka beatboxing), while geometric shapes were used to embody the character of these sounds in visual form – https://vimeo.com/137991974

Nathan Selikoff website links to a number of works – http://nathanselikoff.com/

Audiograph listens to its surroundings and translates sound into light, regardless of the sound’s origin. The tone of human voices, music, traffic, water, and the other sounds found in the city are all different, and will all appear differently on the projected clock face.

 Four Dimensions is a single movement work for orchestra lasting under seven minutes that incorporates surround sound electronics, solo electronic wind instrument (EWI), and real-time visuals – https://youtu.be/N_d8pMxm8Ns
Kimono (Dwaynie East) – 
http://www.surfaces2016.com/wp-content/themes/onetone/js/html5.js
Dwanye East  – is a projection mapped experience that explores the juxtaposition of digital media with the traditional garment, through a variety of unique animations. The aim is to create the illusion of kinetic fabric by meticulously mapping to life-size kimonos specially made for this project – https://vimeo.com/154764946

 Laser Image Sythesizer – Aron Bacs – www.laserlumia.com – enables the creation of flowing structured or abstract laser image artwork using custom software and hardware.  The images are displayed using an 800mW (0.8 Watt) RGB (Red Green Blue) laser projector.  Unlike video where the image is formed via a raster (an array of vertically stacked lines), the laser projector is a vector format such that when a shape, arc, or line needs to be drawn, it can do so without the constraints of resolution typically associated with video.  Lasers enable images to be pure saturated colors, or pastels, and/or vibrant color mixes.  The vector format of the imagery flows similar to a stream of never ending geometric patterns – http://www.laserlumia.com/video-gallery/

22 March 2016

The course has focused on Wekinator exercises which I have not engaged.

3 March 2016

Bibliography recommended by instructor:

1) Data Mining, by Witten, Frank & Hall: http://www.cs.waikato.ac.nz/ml/weka/book.html This is a pretty friendly introduction without too much math, full of practical advice on machine learning and data mining. It’s great for beginners, and can be used hand-in-hand with the Weka software (which Wekinator is built on)

2) Christophe Bishop’s Pattern Recognition & Machine Learning: http://research.microsoft.com/en-us/um/people/cmbishop/prml/ This is a much more in-depth treatment of machine learning, with fantastically clear but also mathematically rigorous explanations for how different algorithms work and what learning means. Requires some comfort with calculus, linear algebra, and probability, so is good for people who are e.g. mid-level undergrads in the sciences or engineering, or beyond.

3) Artificial Intelligence: A Modern Appraoch by Russel & Norvig: http://aima.cs.berkeley.edu/ This is a common choice for an introductory university textbook on AI and machine learning. It covers a LOT of AI topics, so it isn’t as in-depth as the Bishop book, however it’s a really solid and broad introduction to the field.

Exploring Soundation – https://soundation.com/learn/videos

 

My 1st experiment – https://soundcloud.com/ida-brand-o/mixdown-1-soundation 

 

My 2nd experiment – https://soundcloud.com/ida-brand-o/mixdown-2-soundation

 

2 March 2016

A peer introduced a discussion on Neural art and I decided to explore the theme. I found a video on Deep Neural Networks – https://youtu.be/M2IebCN9Ht4 – about how objects are recognized by humans and how the same objects are recognized by the machine – completely different images.

An interesting TEDx on the Art of Neural Networks, Mike Tyka – https://youtu.be/0qVOUD76JOg – Did you know that art and technology can produce fascinating results when combined? Mike Tyka, who is both artist and computer scientist, talks about the power of neural networks. These algorithms are capable to transform computers into artists that can generate breathtaking paintings, music and even poetry.

Another peer shared a link for a tutorial – https://github.com/jcjohnson/neural-style 

Brad_Pitt_into_Picasso_portrait

It came to my mind another tool with typography purposes but that also transforms pictures into different painting styles – http://www.picturetopeople.org/online-text-logo-design-generators.html 

apple images 2

Reminding other generator tools

http://www.malinc.se/m/ImageTiling.php

the end collage

http://deepdreamgenerator.com/

dreamlike generator

Trying Lunapic for animated Gif

free images

 

imageedit_3_3885902982.gif

imageedit_12_9961376912

29 Feb 2016

DeepDream: The art of neural networks – http://grayarea.org/event/deepdream-the-art-of-neural-networks/

Art in the age of machine intelligence – https://medium.com/artists-and-machine-intelligence/what-is-ami-ccd936394a83#.1wcjmt4rm 

25 Feb 2016

An interesting article was shared – Machine learning:  Deep Learning in artistic context- https://medium.com/machine-intelligence-report/machine-deep-learning-in-an-artistic-context-441f28774bcc#.k7hl1k9oz

24 Feb 2016

An open source software «GRIDS»(Topographic drum sequencer) was shared – http://mutable-instruments.net/modules/grids

«Grids’ “brain” is a map of the typical drum patterns used in (mostly electronic) music, laid out by similarity, trained on a large corpus of drum loops. The module can smoothly interpolate and navigate from one pattern to the other, at the whim of a knob move or a CV.»

22 Feb 2016

Wekinator Tutorials – http://www.wekinator.org/videos/

14 Feb 2016

The exercises with Wekinator have started but I have not been engaged.

I have started a new Pinterest board with  Arduino and Makey Makey projects and continued to feed Pinterest Arte Digital e Multimédia .

9 Feb 2016

The videos in Mogees website present several demonstrations and applications of the sensor and software interacting with several types of objects – http://mogees.co.uk/video/ and its Youtube channel also comprises several videos – https://www.youtube.com/channel/UCjVMZp2NSy0lCHBdej99Bjw

Uma apresentação do Mogees na Euronews (PT – https://youtu.be/Tx1jsloZyTM

4 Feb 2016

The course is guided by a single tutor and I’ve listened to the several videos for this first week.

I’ll be more of an observer than a producer since I have no programming or musical background.

I’ve participated in the introductions and discussions and some peers have shared their own works, like these sound experiences – https://youtu.be/4zN7Pjwjcp0, these interaction between body movement and sound – https://youtu.be/v4jz_7PDVfo and these great art installations – http://cuppetellimendoza.com/

A commercial software was introduced MOGEES – http://mogees.co.uk/ which seems exciting as explained in this tutorial – https://youtu.be/Eq7h811wEJM and how it interacts with GarageBand – https://youtu.be/75G1NxLW-Bk 

I have revisited JamStudio which is now a commercial product distributed with another music software – Ignite – http://www.jamstudio.com/Studio/index.htm . I always found this simple software very entertaining. Even if it allows only a try for free.

3 Feb 2016

Machine Learning for Musicians and Artists is an open online course promoted by the University of London in a new platform Kadenze – https://www.kadenze.com/courses/machine-learning-for-musicians-and-artists/sessions/introduction 

Resources required for the course:

  • Materials: Ideally some sensors (could be as simple as a joystick, a webcam, a microphone, or a smartphone with TouchOSC installed, etc.)
  • Equipment: Computer with installation privileges.
  • Software: Wekinator++ – http://www.wekinator.org/kadenze/ 

Some text-based or visual programming background (e.g., an introduction to ChucK, Processing, Max/MSP, PD, or some other environment) is strongly recommended; without this, students will be much more limited in their ability to experiment with the course material. No prior knowledge of machine learning, mathematics, or other topics is required.

Syllabus:

Session 1: Introduction (February 3, 2016)
What is machine learning? And what is it good for? We’ll introduce a variety of artistic, musical, and interactive applications in which machine learning can help you create new things.
Session 2: Classification, Part I (February 17, 2016)
In this session, we’ll cover the basics of classification, which can be used to make sense of complex data in a meaningful way. We’ll look at two classification algorithms: nearest-neighbor and decision stumps. You’ll be introduced to the Wekinator, a free software tool for using machine learning in real-time applications.
Session 3: Regression (February 24, 2016)
We will discuss the fundamentals of regression, which can be used for creating continuous mapping and controls. We’ll explore the use of linear regression, polynomial regression, and neural networks to create new types of interactions. You’ll gain hands-on practice exploring regression algorithms and starting to apply them to build your own systems.
Session 4: Classification, Part II; Design Considerations (March 2, 2016)
In this session, we’ll take a deeper look at what it means to build a good classifier, and we’ll explore some common and powerful classification algorithms, including decision trees, Naive Bayes, AdaBoost, and support vector machines. We’ll also dig deeper into an exploration of how learning algorithms can be integrated into your own work most easily to achieve your desired outcomes. You’ll get a chance to explore these new algorithms and continue to work them into your own projects.
Session 5: Sensors And Features: Generating Useful Inputs For Machine Learning (March 9, 2016)
Machine learning makes it easier and more fun to work with all sorts of real-time sources of data, including real-time audio, video, game controllers, sensors, and more! We’ll talk about good strategies for making sense of the data you’ll get from different inputs, and for designing feature extractors that make machine learning easier. We’ll be encouraging students to develop their own feature extractors and share them with each other!
Session 6: Working With Time (March 16, 2016)
In this session, we’ll talk about algorithms that have been specifically designed to help you make sense of changes in data over time. Rebecca will dive into dynamic time warping, and guest lecturer Baptiste Caramiaux will discuss Gesture Variation Follower, an algorithm designed with the arts in mind. You’ll continue to get plenty of opportunities to apply temporal modeling algorithms to real-time data analysis.
Session 7: Developing A Practice With Machine Learning; Wrap-Up (March 23, 2016)
Guest lecturer Laetitia Sonami will give a masterclass in which she discusses the way machine learning fits into her own work building new musical instruments, and Rebecca will discuss practical tools, boos, and resources you can access for furthering your work in this field.