bhe-aim-imt’s Website
Welcome to bhe-aim-imt website! Relying on The Key Laboratory of Brain Health Intelligent Evaluation and Intervention of the Ministry of Education, our lab is dedicated to solving the problem of intelligent audio intervention and contributing to the advancement of the discipline and society. We welcome people from all areas to visit our website to know more about our research achievements, team members and other contents. We look forward to working with you to explore the mysteries of science!
Research Directions
Our lab’s work centers on three synergistic research directions, forming a complete chain from foundational alignment to dynamic intervention and content generation:
1. Multimodal Semantic Alignment
This direction focuses on aligning the deep semantics between audio signals and brain signals (e.g., EEG) to provide core support for precise and personalized intelligent interventions. We emphasize exploring cross-modal representation learning, semantic embedding, and alignment techniques, laying a solid foundation for subsequent intervention strategy generation.
2. Reinforcement Learning
This direction specializes in using reinforcement learning (RL) to dynamically optimize audio intervention strategies, aiming to tailor and adjust the most effective intervention plan in real time for each user. The ultimate goal is to build a closed-loop interactive, continuously learning, and maximally effective intelligent audio intervention system that is both personalized and precise.
3. Generative Models
This research direction uses advanced generative AI to create new and helpful therapeutic audio. We work with different modern techniques, such as generative adversarial networks (GANs) and diffusion models. Our goal is not only to make the generated audio clear and comfortable to listen to, but also to test and improve its healing effects through scientific methods. We aim to provide practical and new solutions for brain health.
Highlights
Our Research
Our new paper entitled “Enhancing Emotion Regulation in Mental Disorder Treatment: An AIGC-based Closed-Loop Music Intervention System” has been accepted and published online by the IEEE Transactions on Affective Computing (IF-2023: 9.6).
Our Team