Armonías desiguales

Data Sonification / Sound installation

Artist: Nicolás Melmann

Collaboration : Fabian Sguiglia

Research Collaborators:

Economists Martín Burgos, Nadia Shuffer, Emiliano Pazos, the psychologist Matías Gallipoli, and the artists Nicolás Bernez, Fabián Sguiglia and Emilio Marx.

Fragment of the piece

melmann · NICOLAS MELMANN – ARMONÍAS DESIGUALES

INSTALLATION TEASER
ARMONIAS DESIGUALES PREMIER – SETZKASTEN WIEN

PROJECT

Armonías  Desiguales is a generative musical piece, created through data sonification, it i was developed during the residences AIR – Artist in residence Niederösterreich (Austria) and Chateau Ephemere (France) in collaboration with the Master of Sound Art (University of Barcelona) during March and September of 2020. Focusing its attention on the inflationary phenomenon in Argentina (where prices grow around 50% annually), a device created in Max Msp and openFrameworks transforms indicators of the Argentine economy into sound materials: inflation index, poverty, unemployment, salary, devaluation, prices Specific products of the basic basket make up this complex socio-political framework. These data are extracted from official state sources and from informal sources such as prices in supermarkets, becoming variables used to generate music and automate musical parameters.

The project questions an economic model based on social inequality, a machine to manufacture poverty that leaves 50% of the Argentine population living in marginality, where, for example, a monetary devaluation brutally reduces your economic capacity from one second to another. your ability to survive and your reality is instantly transformed as a tonal modulation, it confronts us with another reality and where this economic system has high implications on the character of people. It tries to make a problem audible and in turn transform it into a sound work, loaded with drama and creating atmospheres.The constant changes and transformations of reality are represented through tonal and timbral transformations

Each variable is assigned to an instrument (potentially strings, clarinet and percussion) using samplers in Ableton live, Max translates lists of these numerical values ​​into musical notes.

The analyzed time periods can vary from a century: exchange rate from 1914 to 2019, up to 3 months: price of bread for January, February and March 2020 from Carrefour supermarkets (only in the month of January the price was modified 21 times). 

The narrative nature of the work is given in that different moments in history are seen reflected through sound events, (eg: in each devaluation a tympani sounds). Each variable follows its own cycle, the repetition of the cycles of different durations thus generate phase shifts and infinite harmonic combinations between the different instruments. The rise and fall of musical notes will be determined by the rise or fall of the economic factor (if the price rises, the note rises in register, if unemployment falls, the note will fall in register). In parallel, the timbre and behavior of the instruments will be modified by some parameters of the database and others obtained in real time from the reports of the Central Bank of the Argentine Republic.

MAX MSP PATCH – TEXT TO MIDI

Timbre transformation  / Parameter mapping :

Exchange rate, inflation and prices of the basic basket will be related one by one with pitch, in order to present them in a direct and easily perceptible way. Other data, both included in the database and obtained in real time through the API of the Central Bank of the Argentine Republic, will be used to modify the behaviors and articulations of the sampled acoustic instruments. To work on the behaviors, we have developed a Max for Live Device that uses natural neighbor interpolation to navigate parameter combinations.

This device allows us to store combinations of values for parameters on Ableton Live’s interface and to smoothly interpolate them over time. To map the data to the sampled articulations for each instrument, we have developed a corpus based concatenative synthesizer that organizes samples using a machine learning algorithm (t-distributed stochastic neighbor embedding). This synthesizer performs dimensionality reduction on the sample’s features (mainly on the output of a frequency analysis) allowing us to use the incoming data to browse different recordings. Both devices, developed by the presenters, aim to link the incoming data with regions of similar parameters, allowing gradual modulation in long times on these parameters.

Corpus based concatenative synth
Corpus based concatenative synth
Natural neighbor interpolation
POSSIBLE SCORE OF THE PIECE

Linked work and data source «Mirá que promos» Nicolás Bernez, in the visual sequence we see how inflation works in relation to the price of the empanada for 17 years click here

ARGENTINE PESO – US DOLLAR RATIO 1914 – 2020 CLICK AQUÍ