A general overview of my research.



Taras Kucherenko, Dai Hasegawa, Gustav Eje Henter, Naoshi Kaneko, and Hedvig Kjellström. Analyzing input and output representations for speech-driven gesture generation. International Conference on Intelligent Virtual Agents (IVA ‘19), Paris, July 02–05, 2019

Code is publicly available at Github



Below is a demo applying that model to a new dataset (which is in English). To reproduce the results you can use our pre-trained model


Taras Kucherenko, Jonas Beskow and Hedvig Kjellström. A neural network approach to missing marker reconstruction in human motion capture. arXiv preprint arXiv:1803.02665 (2018).



The system from above could also do Denoising. Code is publicly available at Github