Research Catalogue

This exposition reflects on how machine learning can be integrated with algorithmic composition and live coding to expand digital music creation. The research examines how ML-driven sound analysis, training data, and interactive models reshape compositional workflows. By viewing machine learning as an interpretative and generative process rather than a mere tool, this project challenges conventional boundaries between data gathering, system design, and artistic practice. The discussion is framed through experimental approaches that merge sound synthesis, live coding, and model training, questioning how algorithmic systems can act as both agents of composition and reflective mirrors of musical intention. Through the interplay of structured data, generative models, and exploratory workflows, the study situates machine learning within a broader conversation about creativity, computation, and the evolving role of the composer-programmer.

Anatomy of Loss: An exposition

Building-Sound Compositions in (half-)public places: Starting from Graz, six vacant buildings in different European cities were researched as aural architectures and understood and experienced as an integral part of building-sound compositions. Techniques and strategies ​​were developed how sound art can react systematically to site-specific architectual conditions or how these environmental acoustic characteristics can become part of a previously non-existent composition.

In February 1967, a funeral service was held for the Georgian singer and choirmaster Artem Erkomaishvili. As they bore his body from the municipal theatre in Ozurgeti to his family plot in a nearby village, a recording of the Georgian Orthodox rite for the dead emanated from a portable tape player. The voice on the recording belonged to Artem himself – by this time, no one else in officially atheist Soviet Georgia knew the traditional requiem chants. The previous year, Artem had been involved in an experimental recording at the Tbilisi Conservatory, and he later applied the same technique to capture the three-part chants: going back and forth between two tape recorders, he overdubbed his own voice until all three parts were layered together. Artem’s remarkable, six-decade career stretched from the earliest commercial recordings of Georgian folk music in 1907 to the widespread use of consumer tape-recording technology in the 1960s. His chant recordings – both the conservatory project and the private funeral tapes – were made outside of formal channels of music production and distribution, employing amateur equipment and foregrounding the intimacy of the unvarnished voice. In this way, they resemble magnitizdat, the private tape recordings of poetry and song that circulated unofficially in the late Soviet period (Daughtry 2009). Made at a time when sacred music was still heavily censored, Artem’s recordings occupied the grey area between officially sanctioned and explicitly dissident expression. Building on recent work exploring sound in everyday Soviet experience (Cornish 2020) and expanding the discussion of Georgian music beyond the disciplines of folklore and ethnomusicology, I argue that such private practices of listening and recording provided a means for Georgians in the post-Thaw era to grapple with questions of faith, the loss of tradition, the polyphony of a fracturing state, and the afterlife of a single voice.

This article provides an overview and analysis of transformations of the Belarusian political soundscape. Based on the author’s archive of audio recordings made in Minsk and other Belarusian cities in 2016-2023, the article analyzes how protesters and the authorities used voices, noises, and music during the major political protests of recent years. The field recordings became the starting points for a further discussion on the multifaceted role of sound, music, and silence in contesting for urban and political space in Belarus. The “loudest” period in the country’s recent history is then put in a wider context of studying the clashing ideologies of the authoritarian regime and the democratic, pro-independence movement. The study of audio materials is accompanied by participant observations, interviews, and an extensive analysis of Belarusian and international media that reveal various sonic practices used by the country’s and its critics. The primary goal of this article is to address the gaps in studies of the contemporary Belarusian political soundscape and independent music scenes.

Sashko Protyah is a film director and activist from Mariupol, Ukraine. He's a co-founder of Freefilmers, a collective of artists and filmmakers. In his films, he works with topics of memory, otherness, and alienation. Now Sashko is based in Zaporizhzhia and volunteers for IDPs and the Ukrainian army. This interview was taken in February 2024 over email.