Generating Video Game Sound Effects On the Fly
The tiny sounds players barely notice are often the ones that take the most work to create. Within video game sound design, sound effects such as footsteps or impacts require numerous variations across different surfaces and contexts in order to minimize listener fatigue.
This talk introduces a runtime generative-audio prototype capable of creating video game sound effects on the fly while also producing variations of the generated assets. More broadly, it uses games as a testbed to tackle a harder open problem in audio ML, which is getting neural audio generation fast enough for real-time use.
Tsz Chai Alan Liu
Alan Liu is the Lead Researcher at Sonoform Labs, an independent research lab working at the intersection of music information retrieval, bioacoustics, and machine learning. He builds real-time generative audio systems for games, focusing on neural inference under engine latency constraints. Prior to his research journey, he has also studied classical piano for several years.