AutoMashup: Automatic Music Mashups Creation
By: Marine Delabaere , Léa Miqueu , Michael Moreno and more
Potential Business Impact:
Makes music mixes by matching sounds automatically.
We introduce AutoMashup, a system for automatic mashup creation based on source separation, music analysis, and compatibility estimation. We propose using COCOLA to assess compatibility between separated stems and investigate whether general-purpose pretrained audio models (CLAP and MERT) can support zero-shot estimation of track pair compatibility. Our results show that mashup compatibility is asymmetric -- it depends on the role assigned to each track (vocals or accompaniment) -- and that current embeddings fail to reproduce the perceptual coherence measured by COCOLA. These findings underline the limitations of general-purpose audio representations for compatibility estimation in mashup creation.
Similar Papers
Automatic Music Mixing using a Generative Model of Effect Embeddings
Audio and Speech Processing
Makes music sound better automatically.
Automatic Music Sample Identification with Multi-Track Contrastive Learning
Sound
Finds original songs used in new music.
Automatic Music Sample Identification with Multi-Track Contrastive Learning
Sound
Finds music samples in new songs.