Featured Presenters Announced for AES Symposium on Applications in Machine Learning in Audio

First-ever AES event dedicated to Machine Learning will evaluate opportunities and challenges with insights from top industry leaders and researchers

AES ML Symposium 2020

The Audio Engineering Society has announced a host of presenters slated to participate in the AES Symposium on Applications in Machine Learning in Audio, taking place online, September 28 and 29. This first-ever AES event dedicated to the field of Machine Learning will bring together academic audio researchers and practitioners to offer the latest insight to the capabilities of machine learning applications and technologies. Registration is $25 for AES members and $150 for non-members, the latter including one year of complimentary AES membership and access to full member benefits and resources.

Day one of the Symposium will start with opening remarks and the special presentation "Holly Herndon Artist Talk: Machine Learning & the Creative Process.” Herndon is an Electronic/A.I. composer who, over the last few years has developed, raised and taught her A.I. baby named Spawn. In this presentation she’ll discuss her unique approach to A.I., how it feeds her creative process and the future of A.I. research. Events continue with the presentations “Advances in Breakbeat Analysis, Synthesis and Rhythm Transformation” with Jason Hockman, sound artist, record label owner and associate professor of audio engineering at Birmingham City University and followed later with a session on “Audio Content Analysis” with Alexander Lerch of the Georgia Institute of Technology. The day concludes with the sessions “Machine Listening as a Generative Model: Machine Learning for Music Composition” hosted by University of California, Santa Cruz-based composer and researcher David Kant, and “Neural Audio Synthesis for Music” with Jesse Engel of Google Brain.

Day two digs further into machine learning and related technologies, beginning with “Current Trends in Audio Source Separation” with Fabian-Robert Stoeter of Inria and Stefan Uhlich of Sony R&D Center, followed by “Deep Learning Approaches to Multitrack Mixing” presented by JT Colonel and Christian Steinmetz from Queen Mary University of London. Other featured day two sessions offer “Cognitive Audio”: Enabling Machine Learning Systems with an Understanding of How We Hear,” presented by MIT’s Ishwarya Ananthabhotla, and a look at “Using Machine Learning to Improve Voice Recording, Remix Music and Transcribe Melodies” with Bryan Pardo, co-director of the Northwestern University HCI+Design institute and head of its Interactive Audio Lab. The final session before the Symposium’s closing remarks will gather industry leaders for the in-depth panel discussion “The Challenges of Bringing Audio Products with AI to Market," featuring moderator Jay LeBoeuf moderator with special guests Elias Kokkinis of Accusonus, Drew Silverstein of Amper Music, Maya Ackerman of WaveAi, and Julian Parker with Native Instruments. Further scheduled events include special breakout sessions and time for social interaction among attendees.

www.aes.org