Audio Processing
Comprises Speech Processing and Music Information Retrieval :
|
Music : We design models to learn and represent harmonic structures in music, combining deep learning with music theory. Our work enhances both the analytical and creative capabilities of neural networks across various music information retrieval tasks. |
|
|
Speech Anonymization : We develop inclusive, emotion-preserving speech anonymization techniques that protect speaker identity while maintaining clarity and naturalness. By combining digital signal processing with deep learning and perception-inspired training, our methods enable secure and human-aware use of voice data across diverse speakers and real-world applications. |