I was a PhD student with the PERCEPTION team INRIA Rhône-Alpes France under the supervision of Dr Radu Patrice Horaud and Dr Laurent Girin. My research interests enfold the use of machine learning for audio signal processing and 3D computer vision.
contact: dkounadis at gmail dot com
Joint Source Separation and Diarisation of Convolutive Speech.
Separation of Mixtures with Moving Sound Sources.
An Inverse-Gamma Source Power Prior Model for Audio Source Separation.
3D Point Cloud Alignment (also my Master’s in Greek).
Publication D. Kounades-Bastian, L. Girin, X. Alameda-Pineda, S. Gannot, R. Horaud, “A variational EM algorithm for the separation of moving sound sources”, IEEE Wksp. on Applications of Signal Process. to Audio and Acoustics, New Paltz, NY, 2015 was presented with the best student paper award.