Nicola Privato

University of Iceland, Cultural Studies, Graduate Student

Followers

Following

Co-authors

Public Views

Sreekanth Gopi

Franco Caspe

maziar kanani

University of Art (Tehran)

Kelsey Cotton

Chalmers University of Technology

margaret schedel

SUNY: Stony Brook University

Philippe Pasquier

Çağrı Erdem

University of Oslo

Franziska Schroeder

Queen's University Belfast

Pierluigi Bontempi

Università degli Studi di Padova

Al Hoberman

Interests

Uploads

Papers by Nicola Privato

Notochord Arcs & Scrambled Signals

Zenodo (CERN European Organization for Nuclear Research), Aug 29, 2023

Download

Mouja: Experiencing AI through Magnetic Interactions

Agential Instruments Design Workshop

Zenodo (CERN European Organization for Nuclear Research), Sep 7, 2023

Physical and gestural musical instruments that take advantage of artificial intelligence and mach... more Physical and gestural musical instruments that take advantage of artificial intelligence and machine learning to explore instrumental agency are becoming more accessible due to the development of new tools and workflows specialised for mobility, portability, efficiency and low latency. This full-day, hands-on workshop will provide all of these tools to participants along with support from their creators, enabling rapid creative exploration of their applications a musical instrument design. Description The primary goal of this workshop is to serve as a public-facing, beginner-friendly, hands-on and engaging follow-up to the Embedded AI For NIME [1] workshop presented at NIME '22, where participants can arrive with no prior knowledge of instrument making and finish the workshop by performing on an "agential" instrument of their own making. The previous workshop challenged researchers to address the "technical challenges and higher-level design constraints" associated with making AI/ML techniques work for high performance real-time music interaction. In the intervening period, exciting new tools have started to emerge which are approaching readiness for exploration by non-experts and curious communities like AIMC. In turn, the researchers who are creating these tools need vital feedback to inspire further iteration, refinement, and dissemination to ultimately achieve an open and inclusive ecosystem. Thus, we propose this workshop as a critical step in putting the state-of-the-art into the public's hands. This time however, the scope will not be limited to just embedded systems, and will instead reflect the AIMC theme of interactive performance systems. Itinerary We expect to be able to support between 15-20 participants in this workshop. The full-day workshop will be split into two sessions; hacking and playing. The hacking session will be a hands-on tutorial and tour of various tools and approaches from the workshop organisers, and through this guided process the participants will start to develop their own simple instruments. The playing session will then focus on developing the musicality of these instruments, ending with demos, improvisations and a final discussion. The full itinerary will be as follows: Welcome/Introduction 10-15min intro talk from the organisers and group introductions Session 1: Hacking Short tutorials covering Sensors, gestures and signal processing Acoustics, sound, synthesis, feedback, actuation Mapping, training, refining Comfort break

Download

A Context-Sensitive Approach to XAI in Music Performance

arXiv (Cornell University), Sep 4, 2023

The rapidly evolving field of Explainable Artificial Intelligence (XAI) has generated significant... more The rapidly evolving field of Explainable Artificial Intelligence (XAI) has generated significant interest in developing methods to make AI systems more transparent and understandable. However, the problem of explainability cannot be exhaustively solved in the abstract, as there is no single approach that can be universally applied to generate adequate explanations for any given AI system, and this is especially true in the arts. In this position paper, we propose an Explanatory Pragmatism (EP) framework for XAI in music performance, emphasising the importance of context and audience in the development of explainability requirements. By tailoring explanations to specific audiences and continuously refining them based on feedback, EP offers a promising direction for enhancing the transparency and interpretability of AI systems in broad artistic applications and more specifically to music performance.

Download

The Historically Informed AI: Johann Sperger through time and bass

Zenodo (CERN European Organization for Nuclear Research), Aug 29, 2023

Johann Sperger (1750-1812) was a composer and contrabass virtuoso. Even today, Sperger holds the ... more Johann Sperger (1750-1812) was a composer and contrabass virtuoso. Even today, Sperger holds the record for composing the largest number of contrabass concertos. In his practice, Sperger thoroughly explored the technical and expressive boundaries of his instrument in a dynamic process of reciprocal negotiation between the performer's and instrument's agencies. At the time, Sperger's explorations of the highest register of the

Download

AI Hauntology and the Hauntographic Method

by Nicola Privato and Thor Magnusson

xCoAx 2024 12th Conference on Computation, Communication, Aesthetics and X , Fabrica, Treviso, Italy, 2024

The discourse around creative AI is populated by spectralities and otherworldly presences. Some o... more The discourse around creative AI is populated by spectralities and otherworldly presences. Some of these arise in the political and ethical issues that the technology brings forth, while others haunt the works of artists and designers. This tendency towards the eerie and uncanny, emerging also in my practice, echoes the aesthetics and methods of an artistic movement known as sonic hauntology. In this paper, I explore Derrida's and Fisher's notion of hauntology as an epistemic framework questioning the limits of the metaphysics of presence. I then apply this paradigm to creative AI, and elaborate on the possibility of AI's inherent hauntological potential, arguing that the hauntological in AI arises from the disjunctures that the technology brings forth as it operates with and within the culture. Finally, I introduce AI hauntography, a research methodology combining artistic practice and observation to investigate the phenomenological aspects of creative AI as they intersect with the broader sociopolitical discourse.

Download

The Magnetic Score: Somatosensory Inscriptions and Relational Design in The Instrument-Score

Zenodo (CERN European Organization for Nuclear Research), May 15, 2023

With the changes occurring in the dialectics between the composer and the interpreter during the ... more With the changes occurring in the dialectics between the composer and the interpreter during the second half of the twentieth century, the traditional concept of the musical score has undergone an ontological change. As composers began exploring unconventional notational practices and offering to the interpreter a higher autonomy, the locus of the musical information became less defined, at times merging with that of the instrument. In this paper we explore the dual nature of notation both as score and as instrument from the point of view of non-visual methods of representation. We do this by presenting the Magnetic Score, a system for the inscription and generation of sound that relies on permanent magnetic fields. In magnetic scores, the performative gestures emerge out of the interpreter's embodied interaction with the magnetic fields, and the relational design of the inscriptions together with the interdependence of the symbolic and somatosensory layers offer original insights on the role and situatedness of the musical score in contemporary practices.

Download

Magnetic Interactions as a Somatosensory Interface

by Nicola Privato and Einar Torfi Einarsson

NIME, 2023

Thales is a composed instrument consisting of two hand-held magnetic controllers whose interactio... more Thales is a composed instrument consisting of two hand-held magnetic controllers whose interactions with each other and with other magnets produce the somatosensory manifestation of a tangible interface, that the musician generates and shapes in the act of performing. In this paper we provide a background for the development of Thales by describing the application of permanent magnets in HCI and musical interfaces. We also introduce the instrument's sound generation based on a neural synthesis model and contextualise the system in relation with the concept of magnetic scores. We report on our preliminary user study and discuss the somatosensory response that characterise Thales, observing the interaction between the opposing magnetic field of the controllers as a tangible magnetic interface. Finally, we investigate its nature from the perspective of performative posthumanist ontologies.

Download

THE MAGNETIC SCORE: SOMATOSENSORY INSCRIPTIONS AND RELATIONAL DESIGN IN THE INSTRUMENT-SCORE

TENOR , 2023

Download

Scramble Live: Combining LSTM and Markov Chains for Real-time Musical Interaction

Scramble Live is a complementary tool for Scramble , a 1 standalone MIDI software developed in Ma... more Scramble Live is a complementary tool for Scramble , a 1 standalone MIDI software developed in Max/MSP for 2 the real-time generation of polyphonic music [1]. At their core, both tools combine the flexibility of Markov Chains for the generation of pitch transitions with the generalization capabilities offered by LSTM neural networks on the rhythmic and dynamic domains. Scramble performs the analysis of MIDI files, allows for the selection of the LSTM hyperparameters and the training of the neural network through an intuitive user interface, but offers limited real-time interactivity on playback. Scramble Live relies on the models generated by Scramble , but allows for a high degree of interaction with live instrumentalists and with the system itself. To complete its performance-focused capabilities, Scramble Live offers MIDI mapping and synchronization to external clock sources.

The Society for the Diffusion of Useful Knowledge : Tilting (2)

by Nicola Privato and Blackwood Gallery

As quoted by Jordan Mintzer in "Bong Joon-ho Talks True Crime, Steve Buscemi, Unlikely Success of... more

Download

Generative Scores and Data Mining: W.E.I.R.D. Enters the Stage

by Nicola Privato and Alberto Novello, PhD

xCoAx2021 Conference Proceedings, 2021

W.E.I.R.D. is a suite for piano solo combining the practice of data mining with a generative musi... more W.E.I.R.D. is a suite for piano solo combining the practice of data mining with a generative musical score. Data mining is deployed to extract tweets containing a set of keywords related to the sociological perspective on postmodernity proposed by Zygmunt Bauman. Such keywords are used to trigger the generation of a score written in common music notation, to be performed in real time by a soloist. Through the combination of data mining and generative scores, W.E.I.R.D. introduces the influence of a large sample of society on the stage, thus expanding the performance to emerging contemporary social dynamics. The first two movements of the suite are described from a technical and compositional perspective, and two demonstrative videos are presented outlining how the COVID-19 pandemic has influenced their implementation. Finally, a brief outlook of this work within the field of modern aesthetics and a series of possible developments for such compositional approach are proposed.

Download

Scramble Live: Combining LSTM and Markov Chains for Real-time Musical Interaction

by Nicola Privato, Alberto Novello, PhD, and Omar Rampado

Proceedings of the 19th Sound and Music Computing Conference, June 5-12th, 2022, Saint-Étienne (France), 2022

Download

SketchingSonicTrajectories: A IanniX Tool for composing the Electroacoustic Space

by Julian Scordato and Nicola Privato

Proceedings of the 18th Sound and Music Computing Conference, Jun 28, 2021

This paper provides a report on the development of SketchingSonicTrajectories (SST), a software t... more This paper provides a report on the development of SketchingSonicTrajectories (SST), a software tool for sound spatialization that simplifies the user interaction in a specific usage of IanniX graphic sequencer: SST integrates multi-track audio playback, Ambisonics spatial audio processing, and IanniX score and performance management. In addition to a project description and details on the user interface design, several strategies for the composition of the electroacoustic space with SST will be proposed, also with reference to the score examples included in the software package. The suggested exemplification is intended to illustrate the current software features and guide the user in the design and articulation of advanced movements of sound sources to achieve a multi-channel sound projection without getting entangled in software technicalities.

Download

A Creative Tool for the Musician Combining LSTM and Markov Chains in Max/MSP

by Nicola Privato and Omar Rampado

Artificial Intelligence in Music, Sound, Art and Design

A Creative Tool for the Musician Combining LSTM and Markov Chains in Max/MSP

by Nicola Privato, Omar Rampado, and Alberto Novello, PhD

evoMusArt Conference, 2022

Scramble 4 is a standalone MIDI tool developed in Max/MSP 5 for the real-time generation of polyp... more Scramble 4 is a standalone MIDI tool developed in Max/MSP 5 for the real-time generation of polyphonic music, that combines Markov chains and LSTM neural networks. It offers to the performing musician and composer a simplified user interface for the analysis of MIDI files, the creation of models including expressive parameters beside pitch and rhythm, and the interactive control of the generated output.
In this paper we describe and motivate the strategies we implemented for the analysis and representation of the data, and the encoding techniques we resorted to in order to facilitate the detection of low-level relationships whilst saving computational resources. We also describe a representation of pitch and time domains suitable for both the Markov chain and the LSTM modules, and detail the tool’s architecture both from a functional standpoint and from the perspective of the user. We conclude by present- ing the testing results, by discussing the main limitations of the system and how we intend to address them in future iterations.

Download