What is it about?
Impact sounds (e.g., footsteps) are meticulously created in the post-production process to accompany visual cues in the film, game, and AR/VR industries. We investigate the use of neural audio synthesis for generating such sounds with a Realtime Audio Variational autoEncoder (RAVE) model trained on a dataset of over 3,000 impact sound samples for realtime use in a Digital Audio Workstation (DAW) with interpretable controls. This paper won the 2023 Audio Mostly Best Short Paper Award.
Featured Image
Photo by Kelly Sikkema on Unsplash
Why is it important?
Conventional methods of recording or sourcing these sounds are often costly in time and resources. This paper demonstrates the effectiveness of neural audio synthesis in generating realistic impact sound effects through objective evaluation metrics, and provides open source access to the developed tools.
Read the Original
This page is a summary of: A Plugin for Neural Audio Synthesis of Impact Sound Effects, August 2023, ACM (Association for Computing Machinery),
DOI: 10.1145/3616195.3616221.
You can read the full text:
Contributors
The following have contributed to this page
