Having more than 6 years experience as a software engineer, in 2010 I gave a new start to my career by moving to Barcelona to start a Master’s in signal processing and machine learning applied to music.
Then I worked as a Research engineer for the research institute INESC in Porto, where I was involved in a project aiming at modelling groove, and I developed a few real-time tools for audio analysis and transformation.
In 2013 I have started my PhD at Music Technology Group, Pompeu Fabra University, Barcelona as a part of an European FP7 project which was aiming at transforming the way classical music is enjoyed. My PhD thesis concerned separating the audio corresponding to the instruments in an orchestral music mixture. I gave a new flavour to the topic by proposing a deep learning approach.
I have completed internships at Computational Perception Group, Johannes Kepler University, Linz where I started my work on deep learning for source separation, and at Telefonica Research, Barcelona, where I gave a more general flavour to my research career working on catastrophic forgetting in machine learning (ICML 2018 paper)
I code in Python and C++, mainly using open source library and I try to publish everything I do as open source. I aim at making my work transparent and reproducible, so I can foster future development and creativity.