Juan Montesinos

I’m a PhD student at Pompeu Fabra University in Barcelona.

Image credits: Ariel Ramírez
My main research line is focused on audiovisual sound analysis. Audiovisual content leverages on the co-ocurrence of an even in the real world in both, the audio signal and visual signal, that we percieve. This allows as to perform super wonderful tasks like audio inpainting (based on video), cross-model generation, lip reading, video inpaining and many more things!

I’m currently working on Blind Source Separation and singing voice separation (like in) Cocktail Party Problem. I’m also really interested in Graph convolutional networks.

Since I work in audiovisual I do belong to the Music Technology Group and to the Image Processing Group, being co-supervised by the Dr. Emilia Gómez and the Dr. Gloria Haro.

I’m also a python enthusiast You can find projects I’ve done in GitHub.