You are currently logged in as an
Institutional Subscriber.
If you would like to logout,
please click on the button below.
Home / Publications / E-library page
Only AES members and Institutional Journal Subscribers can download
Modern audio production relies heavily on realtime audio synthesis. However, accelerating audio synthesis far beyond realtime speeds has a significant role to play in advancing intelligent audio production techniques. Fast synthesis methods have been used to generate useful datasets, implement audio matching procedures for automatic sound design, and infer synthesis parameters for real-world sounds. In this paper, we present SYNTHAX, a fast virtual modular synthesizer written in JAX. At its peak, SYNTHAX generates audio over 80,000 times faster than realtime, and significantly faster than the state-of-the-art in accelerated sound synthesis. We present SYNTHAX as an open source1and easily extensible API to stimulate and support applications of fast sound synthesis at scale.
Author (s): Cherep, Manuel; Singh, Nikhil
Affiliation:
Massachusetts Institute of Technology; Massachusetts Institute of Technology
(See document for exact affiliation information.)
AES Convention: 155
Paper Number:10680
Publication Date:
2023-10-06
Import into BibTeX
Session subject:
Applications in Audio
Permalink: https://aes2.org/publications/elibrary-page/?id=22261
(1078KB)
Click to purchase paper as a non-member or login as an AES member. If your company or school subscribes to the E-Library then switch to the institutional version. If you are not an AES member Join the AES. If you need to check your member status, login to the Member Portal.
Cherep, Manuel; Singh, Nikhil; 2023; SynthAX: A Fast Modular Synthesizer in JAX [PDF]; Massachusetts Institute of Technology; Massachusetts Institute of Technology; Paper 10680; Available from: https://aes2.org/publications/elibrary-page/?id=22261
Cherep, Manuel; Singh, Nikhil; SynthAX: A Fast Modular Synthesizer in JAX [PDF]; Massachusetts Institute of Technology; Massachusetts Institute of Technology; Paper 10680; 2023 Available: https://aes2.org/publications/elibrary-page/?id=22261