A Framework for Automatic Mixing Using Timbral Similarity Measures and Genetic Optimization
A novel method is introduced for automatic mix recreation using timbral classification techniques and an optimization algorithm. This approach uses the Euclidean distance between modified Spectral Histograms to calculate the distance between a mix and a target sound and uses a genetic optimization algorithm to figure out the best coefficients for that mix. The implementation has been shown to successfully recreate multitrack mixes accurately and may pave the way towards the automatic mixing of novel multitrack sessions based on a desired target sound.
Click to purchase paper as a non-member or login as an AES member. If your company or school subscribes to the E-Library then switch to the institutional version. If you are not an AES member and would like to subscribe to the E-Library then Join the AES!
This paper costs $33 for non-members and is temporarily free for AES members.