Leveraging compatibility and diversity in computer-aided music mashup creation

Abstract We advance Mixmash-AIS, a multimodal optimization music mashup creation model for loop recombination at scale. Our motivation is to (1) tackle current scalability limitations in state-of-the-art (brute force) computational models while enforcing the (2) compatibility of audio loops and (3) pool diverse mashups that can accommodate user preferences. To this end, we adopt artificial immune system (AIS) opt-aiNet algorithm efficiently compute population compatible from recombinations. Optimal result local minima feature space representing harmonic, rhythmic, spectral musical compatibility. objectively assess compatibility, diversity, performance Mixmash-AIS generated compared standard genetic (GA) brute force (BF) approach. Furthermore, conducted perceptual test validate objective evaluation function within capturing enjoyment computer-generated mashups. results show GA stands as most efficient algorithm, AIS outperforms both BF approaches terms diversity. listening has shown significantly captures ( p < . 001 ).