Latent Arias: The Voice After AI
March 4th, 6 pm – 9 pm. Bath Spa University Locksbrook Campus, Bath, BA1 3EL
Join us for Latent Arias: The Voice After AI to explore the future of the voice and performance in an AI future. With artist Jonathan Chaim Reus, Dr. Francesco Bentivegna and Axel Chemla Romeu-Santos of IRCAM Paris.
A voice tells the story of a body, our sonic presence, deeply tied to authenticity and identity. Our voices are our acoustic footprint, placing us in relation to spaces, each other and other sounds. Processing or morphing the voice has long been a practice that artists working with sound have used to explore their relationship with their bodies, with gender, technology and speculative futures.
With the emergence of AI, new sonic practices and systems allow artists to expand their voices, singing through datasets that transform and respond to the voice. But with the voice being so tied to our bodies, what implications do AI voices have? What is the future of vocal performance with AI?
We’ll be joined by artist Jonathan Chaim Reus, voice studies, Queering AI researcher Dr. Francesco Bentivegna and Sonic AI Developer Axel Chemla Romeu-Santos – to explore the new practices, technologies and ethics of AI voices.
There will be a bar serving soft and alcoholic drinks and snacks throughout the event. If you have access requirements for attending this event, please let us know in the booking questionnaire while reserving your ticket.
Speakers:
Jonathan Chaim Reus, Artist and PhD Candidate at University of Sussex, Co-Founder of Instrument Inventors Initiative & Nederlands Coding Live
Dr. Francesco Bentivegna, Voice Studies and Queering AI Researcher, University of Bristol
Axel Chemla Romeu-Santos, IRCAM Paris, ACIDS Sonic AI Developer
Host:
Nik Rawlings, Creative Producer at The Studio In Bath
Running Order:
6pm Welcome and Opening Comments
6.10pm Presentation 1: Dr. Francesco Bentivegna
Francesco’s research and practice focusses on embodied and queer approaches to AI, by training their voice to sound like an AI voice; a mimetic and inverted approach to most vocal works engaging with AI. They will present this practice, and touch on their Queering AI collective research project (an IAA Project at UoB), exploring queer methodologies with AI.
6.40pm Presentation 2: Axel Chemla Romeu-Santos – IRCAM ACIDS
Axel Romeu-Santos works with IRCAM including on their ACIDS sonic AI projects. Recently these systems have been used in exhibitions by Holly Herndon at Serpentine, and Bjork at the Pompidou, and have also been developed into a range of artist tools using Max/MSP and VST Plug-Ins, enabling artists and creative technologists to create their own sonic data sets, and to train a system that uses this model to resynthesize and respond to live sonic inputs.
Axel will present IRCAM’s work, talk about developing AI/ML tools with and for artists, and demonstrate their systems live.
7.10pm Presentation 3: Performance by Jonathan Chaim Reus
Jonathan Reus is a transmedia musician, performer and researcher. Their multi-faceted work D̴̼̥̉a̸͂͜ͅd̵͂͑ͅą̶̀͐s̵̯̈́e̶̜͔̍͝t̶̘͎̃s̶̫͙͂, funded by the EU’s S+T+ARTS programme, interweaves live performance and data-driven vocal arts a process of creating and training new sonic AI systems using bespoke, collaborative vocal audio datasets.
Dadasets “aims to create work that playfully, intimately and artistically foregrounds less visible labor and the often extractive and exploitative relationships around voice AI, treating datasets and dataset making as cultural and artistic processes imbued with context, and to build conversations and bring attention to desirable care and ethics around human voices under the concept of “vocal values”.”
Jonathan will introduce their work briefly, before performing a live voice work using a bespoke AI voice system.
7.40pm – 10 minute Break / Reset for Panel
7.50 – 8.10pm – Panel Discussion & Q&A – Chair Nik Rawlings (30 minutes)
8.10 – 9pm – Screening of Works & Demonstration of IRCAM ACIDS machine learning system
Axel Chemla Romeu-Santos will be offering an interactive demonstration of IRCAM’s ACIDS machine learning system, while we’ll also be screening documentation of Jonathan Chaim Reus’ works with AI and vocal performances in the Bleachers theatre.