Total Generation Demo (Slakh)

MGE-LDM: Simultaneous Music Generation and Extraction via Joint Latent Modeling

Home Partial Generation (Slakh) Source Extraction (Slakh)

Total Generation Samples

# MSDM[1] MSG-LD[2] MGE-LDM (ours)
#1 Mel Mel Mel
#2 Mel Mel Mel
#3 Mel Mel Mel
#4 Mel Mel Mel
#5 Mel Mel Mel
#6 Mel Mel Mel
#7 Mel Mel Mel
#8 Mel Mel Mel
#9 Mel Mel Mel
#10 Mel Mel Mel
#11 Mel Mel Mel
#12 Mel Mel Mel
#13 Mel Mel Mel
#14 Mel Mel Mel
#15 Mel Mel Mel
#16 Mel Mel Mel
#17 Mel Mel Mel
#18 Mel Mel Mel
#19 Mel Mel Mel
#20 Mel Mel Mel
#21 Mel Mel Mel
#22 Mel Mel Mel
#23 Mel Mel Mel
#24 Mel Mel Mel
#25 Mel Mel Mel
#26 Mel Mel Mel
#27 Mel Mel Mel
#28 Mel Mel Mel
#29 Mel Mel Mel
#30 Mel Mel Mel

[1] Giorgio Mariani, et al., "Multi-Source Diffusion Models for Simultaneous Music Generation and Separation."
ICLR 2024. [Paper]
[2] Tornike Karchkhadze, et al., "Simultaneous music separation and generation using multi-track latent diffusion models."
ICASSP 2025. [Paper]