The following research projects are currently active at MARG:

Hear what you imagine: cross-modal audio synthesis/transformation using generative adversarial network
Auto-regressive Generative Adversarial Network for Singing Synthesis and Evaluation
AcoustoSleepMask: Development of smart sleep mask using acoustic biomarker and machine learning
Urban Data Science Laboratory Project for Future City with Data Foundation
Development of Abnormal Sound Detection Algorithm for Home Appliances using Deep Neural Network
Development of Automatic Lyrics Synchronization Algorithm Using Deep Neural Network
Synthesis of Emotionally Expressive Music Performance Using Machine Learning
Real-time sound analysis technology development
Baby sound database collection and baby condition monitoring using machine learning
HRTF personalization using generative neural networks

The following research projects were carried out at MARG:

Music Recommender Systems
Automatic Music Accompaniment
Smart Interactive Music Edutainment
Tinnitus Therapy based on Music and Brainwave Analysis
Development of Voice-based Diagnostic Marker for Depression
Development of Obstructive Sleep Apnea (OSA) Screening Algorithm through Acoustic Analysis
Blind Perspective
Music Source Separation
Automatic Music Transcription
New Interfaces with Auditory Feedback
Music-lyrics Alignment
Music Visualization for Visually-impaired Persons