MusTBENCH: Benchmarking and Advancing Temporal Grounding in Music LLMs
Research work examining evaluation benchmark, music/audio understanding.
Research
Papers, projects and labs working on AI music.
140 results
N-Singer: A Non-Autoregressive Korean Singing Voice Synthesis System for Pronunciation Enhancement
N-Singer: 한국어 발음 향상을 위한 비자기회귀 가창 음성 합성 시스템
Preprint proposing a parallel Korean singing-voice synthesis system.
NEA (New Electronic Assistant), un sistema para la co-creatividad computacional musical
Peer-reviewed Spanish-language article describing the NEA co-creative music system.
NNSVS: A Neural Network-Based Singing Voice Synthesis Toolkit
Paper describing the design of the open-source NNSVS toolkit.
Noise2Music: Text-conditioned Music Generation with Diffusion Models
Research work examining music generation, music/audio representation learning.
Notochord: a flexible probabilistic model for MIDI performance
Research paper introducing the Notochord generative MIDI performance model.
Pemanfaatan Musik Buatan AI melalui Deepfake
PEMANFAATAN MUSIK BUATAN AI MELALUI DEEPFAKE: STUDI HUKUM KOMPARATIF INDONESIA DAN AMERIKA SERIKAT
Comparative legal study of AI-made deepfake music under Indonesian and United States copyright law.
Piano & AI
UPC project developing a real-time improvisation loop between a pianist and an AI system.
PJS: phoneme-balanced Japanese singing voice corpus
Research paper describing the PJS corpus and its phoneme-balance and licensing design.
Pop2Piano : Pop Audio-based Piano Cover Generation
Research work examining music/audio AI research.
Posthuman Music
Five-year Orpheus Instituut research project examining AI and emerging technologies in music practice.
Posthuman Music
Posthuman Music: Creative Practices after AI and Blockchain
ERC Advanced Grant project investigating musical creative practice after AI and blockchain.
Probing Token Spaces under Generator Shift in AI-Generated Music Detection
Research work examining evaluation benchmark, music/audio representation learning.
Pushing the Frontier of Full-Song Generation: Hierarchical Autoregressive Planning Meets Flow-Matching Rendering
Technical report on a hierarchical autoregressive planner and full-song flow-matching audio renderer.
Qwen-Music Technical Report
Research work examining music generation, singing voice.
RAVE: A variational autoencoder for fast and high-quality neural audio synthesis
Research paper introducing RAVE for fast high-quality neural audio synthesis.