MAAP LAB ๐ต โ Music AI & Audio Research
โWhy are there so few labs in Korea dedicated to Music AI? We built one.โ
Focus areas:
๐ผ Audio Generation ยท ๐ท๏ธ Music Tagging ยท ๐ฃ๏ธ Voice Conversion ยท ๐ง Transformers ยท ๐จ Diffusion
Mission
Advance the foundations of Music AI through practical research in tagging, generation, and dataset-centric methods โ then share our results openly with the community. โจ
Open Science
We aim to publish at top venues (e.g., ICASSP, ISMIR, AAAI) and release code, models, and datasets whenever possible. ๐ข
Latest News ๐๏ธ
- โ
First project achieved! Submitted 2 papers to a NeurIPS Workshop based on our Music Tagging pipeline and dataset work.
Links to be added ๐
- ๐งฐ GPU resources via university support: NVIDIA A100, A6000, RTX 4090 โ๏ธ
Our Activities ๐ฏ
Project 1 โ Music Tagging (Completed) ๐ท๏ธ
- Built a tagging & augmentation pipeline with CLAP, Beam Search, Stable Audio
- Focus: dataset augmentation/creation for future work
- Targets: short-term word generation โ long-term sentence generation with LLMs
- Outcome: 2 NeurIPS Workshop submissions โ
Links: to be added ๐
Project 2 โ Efficient Music Generation (In Progress) ๐ถ
- Exploring Diffusion & DiT (e.g., Flux)
- LoRA/Adapters to avoid full fine-tuning
- Goal: robust generation for data-scarce genres/instruments/domains
- Roadmap: dataset curation โ baseline reproduction โ adapter experiments โ ablations โ release
Publications & Submissions ๐
- NeurIPS Workshop Submission #1 โ (TBD) โ๏ธ
- NeurIPS Workshop Submission #2 โ (TBD) โ๏ธ
Get Involved ๐ค
Interested in collaborating on Music AI? We welcome discussions on datasets, evaluation, and model design.
Contact: arsol970812@gmail.com โ๏ธ
ยฉ 2025 MAAP LAB โข Built with โค๏ธ for music & AI.