Prediction of High-Level-Descriptors in Game Audio

This project aims to research a new set of algorithms that extract audio features from audio files in sound effect libraries and use machine learning approaches to predict high-level audio descriptors–such as emotional adjectives–to enhance game audio file labelling. This research project will assist our partner organization by providing them with a set of algorithms that can be used for the detection and comparison of high level descriptors within their own sound libraries to differentiate their product from competitors, while at the same time providing an improved experience for their users.

Faculty Supervisor:

Miles Thorogood

Student:

Partner:

Audiokinetic

Discipline:

Computer science

Sector:

Information and cultural industries

University:

The University of British Columbia - Okanagan

Program: