Ñe'ê: Earwitnessing hidden stories from the South

Project conceived and developed by Maar Colasso and Gabriel Vigliensoni.

Check the generative radio portal site.

Ñe’ê: Earwitnessing hidden stories from the South is a generative radio project and sound performance that seeks to unveil stories and historical events from the Latin-American South on the idea of disappearance and memory. By placing together sound objects from distant geographies, times, and cultures in a non-binding way—therefore allowing the interaction of their contexts, memories, and temporal densities—this project aims at enabling the audience to reflect and provide meaning to these events.

Our work combines sound sources from different imaginaries from South America to evoke memories and create a speculative aural world. Rather than the usual linear and monolithic treatment of historical events, in our work the narrative and meaning of those events are constructed in the mind of listeners. This experience is driven by a generative soundscape made up of two layers of sound. In the first, a classic, epic poem from Uruguay is resynthesized and told through a computer model learned from testimonies from human rights violations in Chile between 1973 and 1990. The fabled naïvety of the poem is recontextualized and resignified by the memories and verities encoded in the model. In the second layer, keywords extracted from the poem retrieve natural soundscapes of forests, seas, rivers, lakes, flora, and fauna from an online database. Also, rhythms in compound meter—common in traditional music in the global South but also intrinsic to contemporary music genre rhythms—are decoded from a computer model and rendered in real time. Together, all these elements serve as machinery capable of overlapping the fabric of space and time, where words, soundscapes, and rhythms are resignified and can acquire a new meaning in the audience.

This work is based on the idea that in many South American cosmogonies all entities have a soul. Stars, planets, humans, animals, plants, rivers, and even words have a spirit that energizes their semantic meaning. The Guaraní term that names and inspires the project,Ñe’ê, means word but also means soul. In shamanic rituals, we hear the word Ñe’ê accompanied by music. Musical instruments also have a spirit, they speak and teach. In the Guaraní creation myth, music, singing, and dancing are the activities and portals that help people to encounter and communicate with subtle dimensions, divinities, and their ancestors.

The Ñe’ê: Earwitnessing hidden stories from the South project is a portal to alternate aural universes. These universes are explored throughout a generative radio and a live performance will launch the radio portal. The audience will be exposed to sound sources in layers that will be spatialized in the space simultaneously. With this counterpoint of sound sources, the audience will resignify and transfer meaning from one to the other. Similarly, the generative radio portal will broadcast sound sources in parallel, belonging to two layers: palabra and alma.

• The palabra (i.e., word) layer is driven by words. In palabra, the story of the Charrúa people (Uruguay) is told through the aural lens of other people from South America. The epic poem “Tabaré” (Juan Zorilla de San Martín, 1888) narrates the antagonism of the Charrúa and Spaniard worldviews, with the end result of the Charrúas being defeated and exterminated. “Tabaré” is a foundational but fabled component of Uruguayan history—the story never happened. It has been scientifically proved that the Charrúa genocide was a cultural and not a biological one. To resignify the meaning and provide memories to “Tabaré”, the original text will be style-transferred using a machine learning model trained on a set of historical recordings from the Chilean coup d’etat and its aftermath, archived in the Museo de la Memoria y los Derechos Humanos (i.e., Museum of Memory and Human Rights) in Santiago, Chile. These recordings have memories but are not really recollections, they are an aural testimony of what was happening at a certain time in a certain place. Their scope, however, was not only to Chile. Similar situations were happening in other places in South America at the same time. We will use these recordings to generate echoic memories that will be transferred to “Tabaré”.
• The alma (i.e., soul) layer is driven by nature recordings and rhythms. A custom musical instrument will be used as a radio from the past and the future that will serve as a portal to the sound imaginary of the Charrúas. Nature sounds of mountains, hills, stones, and caves; as well as the clouds, rain, earth, sea, rivers, lakes, and the flora and the fauna will be part of the alma layer. Also, rhythms in compound meter will be central. Instead of being subdivided into multiples of two, these rhythms are subdivided into multiples of three. This binary-normative vision of life and music does not apply here. Many traditional rhythms from Africa and South America follow this meter. However, most contemporary research on rhythm modeling is heavily biased toward popular music in simple meter. This bias is detrimental to the modeling of non-Western music. Compound meter rhythms will be central to alma so that these rhythms are not lost anymore.

Ñe’ê is structured as a non-linear, generative aural story. The generative radio is driven by a metadata pseudo score. Words and concepts retrieved from the ten different Cantos in Tabaré (i.e., sections) are used as metadata keywords that drive different audio scenes. The keywords are used as queries that retrieve sounds from Freesound for each audio scene. Additional arguments such as geolocation data and tags (e.g., “field recording”) are used to further filter the results. Probabilities and randomization will be used to further enhance different combinations of sounds. Like a spiral, it follows a cycle but does not have a start or end. Similar to quantum states, sound elements in the spire have a probability of happening, so the story is regenerated every time.

In the opening ritual, we will use three broadcasting systems in parallel. The performance will make use of (i) an Ambisonics 8-channel surround system, (ii) a set of 4 different FM radio signals that will be broadcasted on site and that can be tuned y the audience by using 40 FM radio receivers, and (iii) the online broadcast of a stereo mix that later one can be used for radio broadcast in Deutschlandfunk Kultur’s Klangkunst and Österreichischer Rundfunk’s Ö1 Zeit-Ton or Ö1 Radiokunst. The artists will play with the sound sources through custom digital musical instruments made in MaxMSP, JavaScript, drum machines, and MIDI controllers. Since many of the sound sources are monophonic or stereo signals, for the performance these will be upmixed and spatialized on site.

In the figure, we show a conceptual plan with the components of the live performance that will celebrate the opening of the generative radio portal. Different modes of participation are highlighted in red, different activities in greed, and content layers in blue. The arrangement of the loudspeakers and the placement of the stage within the performance space are explained in the accompanying document with the technical requirements. Colors in the map of South America indicate the territories of contemporary Uruguay (light blue) and Chile (red). The larger colored territory on the map (green) corresponds to the countries where Guaraní populations currently live.