Speculative Sound Synthesis

Edited Online Journal

David Pirrò, Leonie Strecker, Ina Thomann

Published on 15.12.2025

Technological advances in electronic and computer music have historically opened new creative possibilities while simultaneously introducing constraints and standardized practices that shape how we create and think about sound.

Rather than treating these entanglements as merely problematic, Speculative Sound Synthesis asks a different question: What if we actively renegotiated these usually unquestioned interactions?

ECHO #7 — Speculative Sound Synthesis is edited by David Pirrò, Leonie Strecker and Ina Thomann, with contributions by (in order of appearance): Luca Spanedda, Tom Mudd, Ludvig Elblaus, Gerhard Eckel, Thomas Grill, Angélica Castelló, Patrick Lechner, Marco Döttlinger, Arthur Flexer, Chiara Percivati, Andrés Gutiérrez Martínez, Alejandra María Pérez Núñez, Mia Windsor, Nico Daleman, Jo Kazuhiro, Nobuhiro Masuda, and Daniel Louis Lythgoe.

ECHO is an online journal by the Orpheus Instituut, bringing together thought and practice at the intersections of music, technology, and artistic research.

Read full issue Here

Editorial - ECHO#7: Speculative Sound Synthesis

David Pirrò, Leonie Strecker, Ina Thomann

Introduction

This special issue of the ECHO Journal is a collaboration between the ECHO team and the artistic research project Speculative Sound Synthesis, hosted at the Institute of Electronic Music and Acoustics at the University of Music and Performing Arts Graz.

The project—and this special issue—focuses on the complex interplay between technological capabilities, instruments, and aesthetically driven decisions and processes in the practice of sound synthesis. Historically, technological advances in electronic and computer music have opened new creative possibilities while simultaneously introducing constraints, habits, and standardized practices that have significantly shaped artistic practices in this field. Rather than viewing the entanglement of practices as merely problematic, the Speculative Sound Synthesis project proposes a renegotiation of these usually unquestioned interactions.

Our approach emerges from an artistic practice-one where aesthetic concerns and technological development intermingle, where different methods, modes of working, and thinking about sound coexist in a continuous exchange. We conceive this situation as an oscillation—a movement alternating between two poles: artistic or aesthetic praxis on one side, technological or scientific praxis on the other. By praxis we mean theoretically-conscious action, as distinguished from unreflective practice.

This editorial outlines the conceptual framework guiding both the project and this special issue, exploring how speculation operates as both method and artistic stance in this domain.

Full Text

Online Article

David Pirrò

Article published on the Sounding Future Platform.

Speculative Sound Synthesis is an artistic practice, a philosophical concept, and a research method. It is an approach to sound synthesis that challenges conventional methods of sound creation and interaction by interrogating the fundamental relationship between artistic practice and technology. In this article, I will briefly introduce Speculative Sound Synthesis, present its conceptual background, and examine several artistic case studies.

German Version

English Version

Published on the 13.02.2025

Proceedings

Edit by David Pirrò, Leonie Strecker, Luc Döbereiner and Ji Youn Kang

From September 26 to 28, 2024, we hosted the Speculative Sound Synthesis Symposium in Graz. This intense and insporing three-day symposium brought together 40 artists and researchers who responded to our international call for contributions. The online proceedings, which include both their contributions and the video documentation of their presentations and performances, are available publicly on the symposium’s website at this link:

https://speculative.iem.at/symposium/docs/proceedings/

Paper

Luc Döbereiner and David Pirrò

Sound and Music Computing Conference, 2024

Abstract:

This paper explores the intersection of technical, conceptual, and artistic aspects of an ongoing engagement with machine learning (ML), specifically with generative ML in computer music, emphasizing the dynamics of the learning process over its outcomes. Unlike most approaches that prioritize generating “realistic” instrumental or vocal sounds from large datasets, this work integrates the sound synthesis model directly into the learning system, allowing neural networks to produce sound in real-time by continuously adapting and predicting. This method treats the learning process’s behavior as a generative sound synthesis process, making it perceptible. The research delves into how various parameters, such as learning rate and ridge regularization, influence the generation of sonorous behavior, and speculates on the possibilities of a radical reformulation of standard learning functions with a particular interest in complex, chaotic, and relational properties of adaptive computational processes. The paper outlines the motivation behind this exploration, introduces the “Speculative Sound Synthesis” project as its context, presents two case studies to illustrate the approach, describes several experimental artistic applications, and concludes with reflections on the findings and future research directions.

Full Paper

Live-Electronics

Luc Döbereiner, Ji Youn Kang, David Pirrò, Leonie Strecker

Speculative Sound Synthesis: Synchronization is a live electronic perfor- mance by four interlinked players exploring ways of coupling sound synthesis systems, algorithmically, performatively, and sonically. The performance is part of the artistic research project Speculative Sound Synthesis hosted at the Institute of Electronic Music and Acous- tic in Graz. The project challenges established patterns of interaction between technology and artistic practice. Standardized processes in computer music are probed, destabilized and reshaped through speculative re-questioning, thus allowing new aesthetic potentials for experimental musical practice to emerge. Rather than focusing on making the instruments produce specific results, the performers aim to make their instruments’ material qualities, assumptions, er- rors, and even failures sensible and experienceable. By putting their instruments and their respective developers in interaction with each other, creating a feedback loop of sorts, the performance explores questions of interplay, materiality of digital and analog sound syn- thesis, interaction with algorithms and machine learning, employing combinations of nonlinear oscillators and analog circuits based on fundamental digital components. The instruments flow into each other, opening up new sonic and musical possibilities through their coupling, thus creating a laboratory where they can experiment, manipulate, observe and speculate on different aspects of the artistic practice of sound synthesis they consider crucial.

Link to paper

Workshop

David Pirrò, Martin Rumori, and Leonie Strecker

23 - 28 February 2025

with participants of the impuls Academy 2025

Benedikt Alphart | Ted Apel | Francesco Dal Rì | Emanuele Grossi | Georgios Marentakis | Mike McCormick | Nicolas Speda | Justyna Tobera | Lars Fabian Tuchel composition, electronics

The Speculative Sound Synthesis workshop, set up at the impuls Academy 2025, is addressing sound artists, computer music composers and performers who are committed to critical, speculative approaches at the intersection of art and technology, and who are engaged in practices that question, challenge, criticize, deconstruct, recompose, reformulate, shift, dislocate, endanger or reject established standards of sound synthesis and (music) technology. The workshop is part of the artistic research project “Speculative Sound Synthesis”. The aim of the project is to rethink and recompose the relationship between technology and artistic practice in an attempt to unleash aesthetic potentials of sound synthesis that would otherwise remain unknown or hidden within the technological apparatus. In this aesthetic destabilization of analog and digital sound technologies, speculation is the primary method.

During the workshop, each participant developed their own speculative instrument. As a further step, different approaches towards opening up the instruments for receiving and sending sound and data from and to the other instruments were probed, to ultimately create an interconnected, shared agency when playing the instruments together.

Benedikt Alphart ’s instrument is based around a simulacrum-type “physical model” of a singing dune. Singing Dunes are a rare geophonic phenomenon, in which giant heaps of sand produce booming bass drones reaching up to 120dB in volume. The model allows him to mimic its behavior and push parameters beyond their natural range, using his custom built “Mozzarella”-controller.

Ted Apel ’s algorithm uses a spectral representation of sound to manipulate the phase spectrum of sound for unique time manipulations of other performers’ sounds. The amplitude envelope of the resultant sound is controlled by OSC messages from other instruments.

Francesco Dal Rì ’s instrument is centered around three types of feedback: acoustic, signal, and data. The main software component consists of two identical pipelines, each containing a digital multitimbral synthesizer and a neural network, trained to estimate parameters for the synthesizers to emulate input sounds. Using a hardware mixer as main control interface, the behavior of the system is influenced via multiple feedbacks approaches, achieving both stable and chaotic states.

Emanuele Grossi ’s instrument consists of a granulator with per-grain spatialization and stretched parameter possibilities meant for working with large prerecorded field recordings, in this case adapted to record material and use parameters coming from the other musicians’ instruments.

The NSFW (‘Nuther SuperCollider FrameWork’) is a software developed for highly dynamic and expressive control of digital synthesis and processing algorithms in performance. During the Speculative Sound Synthesis workshop Mike McCormick has developed a new module for the system – the NS_{StripRegressor} – which allows him to control many modules at once through a simple interface by using a neural network to perform regression on higher dimensional parameter data.

The instrument of Georgios Marentakis couples live-camera input from a mobile phone to sound synthesized and played or streamed in real-time by the device. This is done based on a series of user-adjustable mappings that shape the timbre and the dynamics of the generated sound. The intention is to explore serendipitous sonic outcomes and emphasize them through gestural performance.

Nicholas Speda receives sound from another participants’ speculative synth and live-codes complex and chaotic effects for it to pass through. Whenever he evaluates code to make adjustments or additions to these effects, the input source switches to a new (random) participants’ synth, making it impossible to predict the outcoming sound beforehand, thus turning every modification of the code into a speculative endeavor. In addition to this, data from typing out his code (e.g. statistic of used keys) is sent out to other participants to further influence their speculative synths (e.g. use as envelopes).

Justyna Tobera ’s instrument is a sensor that reads hand gestures and processes sound, distributing it across a multichannel system. It allows for real-time interaction, where subtle movements influence the spatialization of sound. At the same time, it generates and transforms visualizations based on movement, creating a dynamic audiovisual experience.

By combining techniques from granular, concatenative, and waveset synthesis, and incorporating cutting-edge linguistics software, the speculative sound synthesis system stack named mund, created by Lars Tuchel, generates rich textures and intricate (a-)rhythmic patterns. The smallest sound elements – the grains, the waves, the phonemes – are layered and combined to reveal a voice hidden within diverse sound corpora.

Workshop

David Pirrò, Luc Döbereiner, Ji Youn Kang, and Leonie Streker

22 March 2024

The workshop will start with the team introducing and sharing the instruments developed as part of the research project. In the following hands-on session, participants will have the opportunity to speculatively explore their own musical instruments (e.g. a block of code, a synthesis tool, an analog circuit, and/or any sounding objects) with the other participants. This practice is not only intended to facilitate exchange and experience of instruments, but also to delve into the speculations that arise when encountering new sonic possibilities. Before diving into the music, each couple of participants will pose speculative inquiries, sparking discussions on the role of speculation in the performance process.

As participants navigate the unfamiliar territory of someone else's instrument, they will uncover new layers of sonic potential and artistic expression. Through collaborative improvisation sessions, small groups will come together to share their experiences, reflecting on the ways in which speculation shapes their artistic process. The project team will guide and take part in this session as well.

Lecture-Performance with Live-Electronics

Luc Döbereiner, Ji Youn Kang, David Pirrò, Leonie Strecker

Tuesday, 21.05.2024, 17:00, IEM CUBE

In this lecture performance, each team member shared their thoughts on the role of speculation in their artistic practice in a speculative form, navigating the landscape between aesthetics, sound, technology and methodology through a speculative lens. They each chose keywords related to their practice, thoughts or views on speculation, which were presented in a strictly timed manner. The entire lecture lasted exactly 82 minutes, including short intermissions during which music was played using interconnected synthesis systems.

Video and Audio Recording