Federico Visi

Tag: Sophtar

  • Wilding AI Lab, CTM 2025

    Wilding AI Lab, CTM 2025

    Next week I will take part in the Wilding AI Lab:

    this four-day lab will assemble a group of participants selected via open call to learn about the application of generative AI in spatial audio, and collectively explore the wilder territories of AI. Shaped as a mix of theoretical and hands-on components, the lab runs 23 – 25 January, and culminates with a public presentation session Sunday 26 January.

    The lab will take place at MONOM, a unique performance venue and spatial audio studio housed at Funkhaus Berlin. It consists in a 3D array of omnidirectional speakers arranged on columns and subwoofers placed underneath the audience, beneath an acoustically transparent floor.

    MONOM (Ph: Becca Crawford, from 4dsound.net.)

    I am pretty excited about experimenting with the Sophtar there, particularly given that each of its eight strings and the audio synthesis machine learning models can be routed to separate output channels, making it possible to place each sound source in a different point in space.

    The Sophtar (Ph: Federico Visi).

    I look forward to meeting the other participants in person. Engaging with a community of practitioners, share skills and ideas, and take part in an open critical discussion on AI- and data-driven tools is crucial to address the rapid changes affecting our cultural landscape. I believe it is also very important to open the process to the public in order to demystify the black boxes, promote critical thinking, spread knowledge, and offer new narratives, so I am excited at the prospect of an open lab on the fourth day. As in Maurice Jones’ words during our first collective Zoom meeting “the design of how we gather is essential.”

    Participants

    Hosts

  • The Sophtar

    The Sophtar

    The Sophtar is a tabletop string instrument with an embedded system for digital signal processing, networking, and machine learning. It features a pressure-sensitive fretted neck, two sound boxes, and controlled feedback capabilities by means of bespoke interface elements. The design of the instrument is informed by my practice with hyperorgan interaction in networked music performance.

    I built the Sophtar in collaboration with Sukandar Kartadinata. I presented and performed with it at NIME 2024, here is the paper from the conference proceedings.

    At IIL I am working on extending the Sophtar with actuators and machine learning models to make it respond to my playing in ways that are not easy to predict yet meaningful and inspiring. In particular, I am working on:

    • an extension that allows the instrument to self-play by means of solenoids,
    • embedding notochord models,
    • per-string filtering for harmonic feedback.

    I see the research and development work on the Sophtar as a way to probe and engage with broader research questions on musical improvisation and co-creativity with machines and algorithms.

    I will present the Sophtar at an Open Lab on Friday, 27th September, 15:00-17:00.

    Playing the Sophtar at LNDW 2024 Photo: Christian Kielmann.

  • Intelligent Instruments Lab, Iceland

    Intelligent Instruments Lab, Iceland

    My family and I temporarily left Berlin and relocated to Reykjavik, where I am going to join the Intelligent Instrument Lab (IIL) at the University of Iceland. There, I am going to work on extending the capabilities of the Sophtar, the musical instrument I designed to address the needs stemming from my artistic practice involving hyperorgans, networked music performance, and interactive machine learning. I built the Sophtar in collaboration with Sukandar Kartadinata and I describe it more in detail in this NIME 2024 paper. I very much look forward to working with Thor Magnusson and the other super talented researchers at IIL, starting in just a few days!