The Walt Disney Company
System for deliverables versioning in audio mastering

Abstract:

Implementations of the disclosure describe systems and methods that leverage machine learning to automate the process of creating various versions of sound mixes using an original sound mix as a starting point. In implementations, a system for automated versioning of sound mixes may include: (i) a component to extract metadata categorizing/identifying the input sound mix; (ii) a component to extract audio features of the input sound mix; (iii) a component that uses a machine learning model to compare the extracted audio features of the input sound mix with extracted audio features of previously analyzed sound mixes to calculate audio features of a target sound mix; and (iv) a component to perform signal processing to derive the target sound mix given the calculated audio features.