Info Mining: A Look at George Tzanetakis' Innovations in Music Classification

(0)

 

Printed scores were once necessary for music listening.  Until the 20th century, each musician playing a symphony would need his own notated sheet music in order to play a piece for every performance.  Today, the bulk of music listening happens through recordings.  Musicians only need to play a song correctly once in order for anybody to hear it anytime, anywhere.

But with the streamlined dissemination of digital music on the Internet, today’s listeners need guidelines for how to consume music just as badly as musicians once needed scores to produce new music.  There is simply too much recorded music for any one person to keep track.  Accordingly, “music discovery services”, which guide listeners through huge libraries of music, are beginning to emerge as a genuine growth industry.

Pandora, a leading music discovery service, famously began its Music Genome Project about a decade ago, a music classification method that numerically rates songs according to a long list of criteria and sorts songs by these “genetic” similarities.  Pandora’s website generates playlist suggestions based on a minimal amount of input from listeners.  Ideally, Pandora automatically can create personally tailored playlists that a listener didn’t have the knowledge or time to create.

Shortly before the Music Genome Project commenced, George Tzanetakis made Marsyas, an open-source toolkit for automatically classifying songs and entire libraries of music, among other applications.  Pandora and Marsyas had similar aims - to intelligently sort music libraries to give listeners a way to find new artists and retrieve other qualitative information about music.  Working at Princeton as a grad student with professor Perry Cook, who wanted to find a way of automatically sorting radio stations, Tzanetakis developed various library-browsing visualizations within Marsyas, including Genre Meter, which can respond live to sound sources and classify them (video demo.)

Pandora has taken off as a large-scale commercial venture, with more competitors like Spotify and Slacker in its wake.  Tzanetakis’ Marsyas has remained known mostly only by academics and computer scientists.  Regardless, Tzanetakis’ work addresses issues of music classification in a more radical and even prophetic way than Pandora: all of Marsyas’ “genes” are completely determined by computer automation.  Tzanetakis’ contributions to the field of Music Information Retrieval (MIR, for short) have helped to push computers toward increasingly delicate interpretations of one of man’s most elusive forms of expression.  Marsyas is available for free download and even has a free user manual.

Though songs in the Pandora database are weighed and sorted by algorithms, a board of experts determines the value of each “gene”.  Recently, a New York Times reporter sat in with a group of Pandora’s experts listening to songs and then opining about how high a song scored in criteria like “emotional delivery”, “exoticism” and “riskiness”; as well as more concrete judgements on tempo, instrumentation and harmony.

By contrast, George Tzanetakis’ approach to music classification is completely automated.  It needs no panel of experts or crowdsourced participants to complete an intelligently made, intuitively browsable library of music.  It works based entirely on the audio signals themselves.  Given merely a library of digital song files, George Tzanetakis’ automated classification techniques algorithmically organize songs according to a variety of criteria and present fun interactive ways to browse and compare music.

 

READ ON »


Pure Data read as pure data (2010) - Nicolas Maigret

(0)

The content of the Pure Data application is read as pure data into sound and pixels (rgb + extrude)

LINK »


Images from Jonathan Zawada's exhibit "Over Time" at PRISM

(0)


Why the Earth is Green, 2010

The exhibition focuses on large-scale landscape paintings whose topographies are derived from graph data. Zawada collected and compared a variety of data series that extrapolate information over time, such as “Marijuana usage among year 12 students vs. CD and Vinyl record sales between 1975 and 2000” or “Value of land per square meter in Second Life vs. Value of land per square foot in Dubai between 2007 and 2009.” The data is then manipulated through a 3D fractal program and the resulting environment becomes a virtual abstraction that mimics a mountainous landscape.

Painted on linen, the landscapes are a response to the “virtual” reality of digital experiences that are highlighted by the intrinsic flatness and surreal color palate. Invoking the robotics hypothesis of the “Uncanny Valley,” the works take on an android quality, a sense of reality but not quite, registering with the viewer as both familiar and dissimilar. This theme carries through to his drawings, juxtaposing the hyper-real with the conceptually abstract and underlining the temporality of human experience.

-- DESCRIPTION FROM STATEMENT FOR "OVER TIME" AT PRISM (DEC 16, 2010 — FEB 12, 2011)

Flight 77, 2010

Earth Movers, 2010

Very Hot Nights, 2010

Land Sale, 2010

Originally via TRIANGULATION BLOG

LINK »


Required Reading

(1)

harwood4-675.jpg
YoHa (Graham Harwood and Matsuko Yokokoji), Coal Fired Computers, 2010.
(Installation view at Discovery Museum, Newcastle, courtesy the artists. Photograph: Louise Hepworth)

This interview follows on from a project called “Coal Fired Computers (300,000,000 Computers - 318,000 Black Lungs)” carried out in Newcastle in spring 2010 for the AV Festival. The project, by Graham Harwood, Matsuko Yokokoji with Jean Denmars involved a means of producing a physical diagram between components in production as they undergo transformations across different kinds of time, politics, matter, knowledge, and vitality. The project found a way of working with such things that was particularly powerful. The interview begins with a discussion of CFC but also moves off into databases and a certain understanding of their material force. One thing we don’t cover in the interview is the detail of the Coal Fired Computers project’s work with miner activists, including the inspirational Dave Douglass. (See information on his memoirs here ). More of this can be found in a booklet about the project here, including links to all the groups involved. The interview was carried out by email in May and June 2010.

-- EXCERPT FROM "PITS TO BITS, INTERVIEW WITH GRAHAM HARWOOD" BY MATTHEW FULLER

MORE »


Following the Lines

(0)

JWinstalledit.jpg
Jeremy Wood's "Mowing the Lawn" Installation View at Tenderpixel

In an era of Google Maps, our first engagements with places are often anticipated by technology. That is, our experience with a place often comes with pre-emptive associations from aerial pictures -- our possible routes predetermined and mapped; personal narratives and exploration are displaced for utility. So, what happens to our individualized explorations in time and space when GPS technology intervenes? This is the inquiry of GPS artist Jeremy Wood’s body of work and his current show “Mowing the Lawn” at Tenderpixel in London.

Lawn2005Scale1-300edit.gif
Jeremy Wood, Lawn 2005 Scale 1:300, 2010

Treating his body like a “geodesic” pencil, his daily routines are documented as lines in space via GPS technology. In turn, Wood’s performative rituals are data visualized as densely packed line drawings and animations. Having spent ten years developing a system for tracking and translating his everyday movements, the resulting pieces are one part drawing, one part diary and one part critique of the technological system’s accuracy/inaccuracy and how that intervention enables/limits our perception of the spatio-temporal.

NineYearsofMowingedit.jpg
Jeremy Wood, Nine Years of Mowing, 2010

While his work ranges from tracking large-scale transatlantic flights (Star Flights, 2008) to tracing and superimposing quotes from Melville onto two meridians in London (Meridians, 2005), in his latest show, Wood focuses on documenting the simple act of mowing the lawn in different intervals of time. Here, Wood emphasizes how banal repetition offers “individual narratives that express a freedom of movement generated from an act of garden maintenance”.

What may be more compelling, though, is how a digital trace can bring to the fore the problems of technology. Looking at Lawn 2005 Scale 1:300, we see multiple lines drawn where a house already exists. In Nine Years of Mowing ...

MORE »


NOWHERE (2005-2006) - Ralf Baecker

(0)

nowhere0.jpg


Nowhere is a three-dimensional milling machine that carves a landscape relief on a 70x70x10cm large block of hard foam. The machine receives a stream of live search requests from the german search engines metager and metager2 (www.metager.de) via the internet. The users search movements erode rivers and canyons on the surface. Search requests that shoot through the internet just for a fraction of a second and generate an answer on the searchers screen, cause the machine to write a constant growing sculpture into the space. The continuous stream of changing search requests defines form and rhythm of this process.

-- DESCRIPTION FROM THE ARTIST'S STATEMENT

MORE »


General Web Content

(3)

For today's General Web Content I have assembled a collection of images that repurpose traditional models of data visualization for humorous/bizarre/illuminating effect. This meme has been around for several years now, first coming into mainstream awareness with the emergence of the overwhelmingly brilliant website "rap represented in mathematical charts and graphs," and continues to be a persistent mechanism for creative expression across the web. (Especially in forums such as b3ta, 4chan, and Something Awful.) The intent of this collection is not to present a best of, but merely to convey a broad overview of the meme. Enjoy.



























































MORE »


Knowledge Work(s): In Search of a Spreadsheet Aesthetics

(1)

I sympathize with the protagonist of a cartoon claiming to have transferred x amount of megabytes, physically exhausted after a day of downloading. The simple act of moving information from one place to another today constitutes a significant cultural act in and of itself. I think it's fair to say that most of us spend hours each day shifting content into different containers. Some of us call this writing.

- Kenneth Goldsmith, 2004

While Kenneth Goldsmith's wry statement about knowledge jockeying is directly discussing the plight of the contemporary author, his comments are useful for thinking about other disciplines. In editing this quote, the word "writing" could easily be replaced by any number of verbs (programming, composing, painting, storyboarding, etc.) as we undoubtedly inhabit an era where creative transposition rather than raw creativity can be enough to drive a project. The ctrl-c clipboard, the layer palette in photo editing software and the flash memory of a microcontroller are all examples of spaces that serve as staging grounds for storytelling and crafting aesthetic experiences — these are interstitial zones where art gestates. Goldsmith clearly doesn't approach the creative process with reverence, and his blasé attitude is an excellent springboard into reading contemporary artistic production in relation to knowledge work. An important question: How might we appropriate this daily activity of "shifting content between containers" as a site (rather than a means) of artistic production? This article will consider the aesthetics of the spreadsheet, and act as the first installment of a series that will engage projects that explore the documents, software, interior architecture and politics of the contemporary workplace.

READ ON »


Frameworks, Crowdfunding, Cassandra and Undocumented Wind Instruments

(0)

Now that everyone is out of business cards and has had enough time to check in to their locative media apps, I think we can begin to make sense of the social and technological deluge that was the South by Southwest Interactive festival. After being deep in a web development hole for the past few months, what I took away from the conference was a rejuvenation of critical, big-picture questioning, a reminder of just how drastically technology is contouring contemporary society and culture and that, ultimately, it is still in our hands to determine the overall shape of things to come. Although a late arrival and scheduling conflicts prevented me from hearing everyone I'd have liked to have heard (Douglas Rushkoff, Gary Vaynerchuk, among many others), I was able to take in most of the keynote speakers and the panels whose subject had some impact or connection to the arts (which were few). Here is a synopsis of the projects, presentations, and people that resonated with me the most.

READ ON »


Interview with Sinan Ascioglu

(1)

drivingthroughiceland.png
"Driving through Iceland” sketch by dotlassie. Winner of Rhizome's Tiny Sketch Competition.

OpenProcessing.org is a site that has built a community around sharing visual coding examples created in Processing. As user number 36, I had the unique privilege of watching the idea take shape, while in a thesis group with Sinan at NYU’s Interactive Telecommunications Program. During it’s first two years of activity, the site has grown to host thousands of user-generated sketches and subsequent conversations between artists / programmers, teachers, and students from around the world. Sinan and I escaped the snow recently at a café outside Washington Square Park to discuss OpenProcessing’s origins, Rhizome’s collaboration with OpenProcessing in the Tiny Sketch competition, and what we can expect for the future. - Tim Stutts

Tim: How did you first come up with the idea for OpenProcessing?

Sinan: I guess the first thing to talk about is OpenVisuals, which was my Master’s thesis project at ITP (Interactive Telecommunications Program, New York University). I was reading Edward Tufte’s books at the time, and I became very interested in data visualization. In the meantime I was also fascinated with the social revolution that was happening on the web, through a class I was taking at NYU with professor Clay Shirky. Before studying with Clay, I didn’t understand Facebook—I didn’t even have a Facebook account. I knew I was missing some concepts, and wanted to understand what was going on. Through his classes, I decided that my thesis would have a social component. I also discovered ManyEyes, a site where users can upload datasets, and choose between different the visualization methods for augmentation. Users comment on each other’s visualizations, and may even suggest other ways of looking at or representing the dataset. What ...

MORE »