AES Symposium on Applications of Machine Learning in Audio

July 17 2020, 09:10
The Audio Engineering Society is inviting academic and industry audio researchers and practitioners to participate in the first AES event highlighting Machine Learning. The online symposium will focus on machine learning as it relates to applications in digital signal processing and audio engineering. The recently announced Audio Product Education Institute is also supporting this important AES symposium, sponsored by iZotope. September 28-29, 2020.

Topics for this online event will include, but are not limited to, practical applications of machine learning in automatic mixing, audio source separation, audio visualization and effect control, sourcing audio data for legal issues, open datasets, etc. Another increasingly popular application field for machine learning is its use for audio capture and recording, including assistive recording in poor acoustic contexts.

The online event will take place September 28-29, and will offer nine hours of programming consisting of pre-recorded presentations with live Q&A and parallel sessions in dedicated breakout rooms. Additionally, each day of programming will conclude with an online social hour. The symposium will be scheduled over two four-hour-days to maximally accommodate participants in different geographical regions.

The event's committee is headed by Jonathan Wyner, AES President-Elect (2020) and currently Education Director for iZotope in Cambridge, MA, with program chairs Christian Uhle (Chief Scientist at Fraunhofer IIS), Gordon Wichern (Principal Research Scientist, Mitsubishi Electric Research Laboratories, Cambridge, MA, USA), and Pippin Bongiovanni (Musicologist and iZotope Content Marketer).

Keynote speaker for the symposium will be composer Holly Herndon. After studying composition at Stanford University and completing her Ph.D. at Stanford University’s Center for Computer Research in Music and Acoustics, Herndon pursued an international career, currently based in Berlin, Germany. In her recent full-length album PROTO, Herndon fronts and conducts an electronic pop choir comprised of both human and A.I. voices over a musical palette that encompasses everything from synths to Sacred Harp stylings.

The 2020 AES virtual event, Applications of Machine Learning in Audio is accepting proposals for parallel breakout sessions. Accepted presenters will submit a pre-recorded video between 5 and 10 minutes long, which attendees of the symposium will have the opportunity to view beforehand. During the session, presenters will be online in an interactive video channel, where they may present more material (e.g., slides) and answer questions from attendees. Deadline for submissions is August 9, 2020

Registration is $25 for AES members, and $150 for non-members (includes 1-year AES membership).
More details here:
related items