Automatic Mixing Through Machine Learning

Automatic mixing is a field of music technology research looking to automate sound engineering processes. The project will look to expand the field, using machine learning techniques to analyze the relationship of the subjective sound engineering techniques of audio engineers. From this, a unique and quantifiable data set will be we gathered in order to represent a sound engineer’s unique ‘sound profile’, which will enable the automatic modelling of different sound engineering styles. The subjective and objective comparison will be made possible by systematic tests will be run to analyze professional audio engineering perceptions and preferences, signal analysis of reputable signal processors, and the mathematical retrieval of music information.

Intern: 
Matthew Boerum
Superviseur universitaire: 
Dr. Richard King
Project Year: 
2014
Province: 
Quebec
Université: 
Partenaire: 
Discipline: 
Programme: