SongCompose-PT
dataset·bilingual lyrics melody pretraining dataset·active
Chinese-English pretraining dataset containing lyrics, melodies and paired lyric-melody data for SongComposer.
Recorded facts
| Official site | https://arxiv.org/abs/2402.17645 ↗ |
|---|---|
| Geography | Mainland China, China |
| arxiv | 2402.17645 |
| access note | Paper documents the corpus; a current public download and license were not located. |
| dedupe risk | low |
| discovery route | Simplified Chinese, Traditional Chinese and English source discovery |
| evidence strength | strong |
| recommended prompt | 06-datasets-and-benchmarks |
| classification rationale | Primary evidence shows material bilingual lyrics melody pretraining dataset relevance to AI music or audio AI. |
Current
| used by | SongComposer: A Large Language Model for Lyric and Melody Generation in Song CompositionSource 1 ↗ VERIFIED high confidence |
|---|
Sources & changes
Checked 10d ago · highhow verification works
Field-level evidence
Public change history
- Status unknown → active
- Official URL unknown → arxiv.org/abs/2402.17645
- Record maintenance · 12 fields updated
Is this yours? Claim this record →·See something wrong? Report a correction →