Masked Audio Generation using a Single Non-Autoregressive Transformer
Research work examining audio generation, text-to-audio generation.
Research
Papers, projects and labs working on AI music.
140 results
MERT: Acoustic Music Understanding Model with Large-Scale Self-supervised Training
Research work examining music/audio understanding.
Mimerska väven
KTH artificial folk-music composition project drawing on the FolkWiki tune archive.
MishMash AI and Creativity research programme
Multi-institution research programme developing and studying co-creative AI, with dedicated music and artistic work.
Missing Melodies: AI Music Generation and its "Nearly" Complete Omission of the Global South
Research work examining music generation, music/audio representation learning.
MLP Singer: Towards Rapid Parallel Korean Singing Voice Synthesis
MLP Singer: 빠른 병렬 한국어 가창 음성 합성을 향하여
Preprint describing a fast parallel Korean singing-voice synthesis architecture.
MMAE: A Massive Multitask Audio Editing Benchmark
Research work examining evaluation benchmark, music/audio editing.
MMAU: A Massive Multi-Task Audio Understanding and Reasoning Benchmark
Research work examining evaluation benchmark, music/audio understanding.
Moises-Light: Resource-efficient Band-split U-Net For Music Source Separation
Paper presenting and benchmarking the Moises-Light source-separation architecture.
Montreux Jazz Archives Secure Research Access
Infrastructure for controlled computational musicology access to the Montreux Jazz audiovisual archive.
Moûsai: Text-to-Music Generation with Long-Context Latent Diffusion
Research work examining music generation, text-to-music generation.
MT3: Multi-Task Multitrack Music Transcription
Research work examining music transcription, music/audio understanding.
MuChoMusic: Evaluating Music Understanding in Multimodal Audio-Language Models
Research work examining evaluation benchmark, music/audio understanding.
MuLan: A Joint Embedding of Music Audio and Natural Language
Research work examining music retrieval, music/audio representation learning.
MUSCLE
Infrastructure for music-related citizen-science listening experiments with current structural support.
MuScriptor: An Open Model for Multi-Instrument Music Transcription
Research work examining music transcription.