SPDMX releases 254K-song open multitrack music dataset synthesized from public-domain scores
umpedronosapato · x · 2026-11-09
The SPDMX paper introduces a listening-guided rendering of public-domain PDMX scores into the largest openly-licensed synthetic multitrack corpus: 253,982 songs, 490,159 mono stems, and 5,469 song-hours at 44.1 kHz under CC BY 4.0, with paired MIDI and PDMX metadata. 34.8% of songs (2,734 hours) offer multitrack supervision, dwarfing real-recording sets like MUSDB18 and enabling source separation, stem-conditioned synthesis, MIDI-to-audio, transcription, and joint symbolic-audio modeling. Corpus, metadata, and the open pipeline are released.
More from Research
- SplatAnything debuts: feed-forward 3DGS reconstruction that generalizes to arbitrary scenes — _akhaliq · 2026-11-10
- Tencent releases TailorRig: 39,073 rigged humanoid meshes for auto-rigging research — _akhaliq · 2026-11-10
- P²-MOPD uses expert weights as parameter priors to halve multi-teacher distillation steps — _akhaliq · 2026-11-10
- KlingAI releases LongBanana: 90-task benchmark for multi-reference image synthesis — _akhaliq · 2026-11-10
- Satellite Image Registration Still Uses Classical Cross-Correlation; Modern Matchers Lack a Benchmark — gabriberton · 2026-11-10
- Epoch AI researcher: most 'algorithmic progress' in frontier AI may come from better data — S_OhEigeartaigh · 2026-11-10