I have participated in several research and industry related projects. During my PhD I have mainly been envolved in the PHENICX project. I have also been envolved in #MusicBricks, with “Sound in Translation” which was the winner of the incubation in the BCN Music Hack Day 2015.
- PHENICX project: The main goal of this project is to make classical music concert performances come alive to broad audiences. I have mainly been working on a melody extraction method, a dataset and visualisation tool for melody extraction on symphonic music.
- The method achieved best overall accuracy in MIREX 2015 on symphonic music, but also on other datasets containing rock, pop, jazz, and R&B music. It has also been evaluated in large datasets such as MedleyDB, achieving significant improvements over the state-of-the-art.
- Results of my research have been integrated into a prototype.
- The dataset has been used in MIREX, and is available for the community.
- The tool allows online visualisation of (score-informed) melody information, including the pitches (related to the notes) and the instruments playing the melody.
- Sound in Translation: This project started in the BCN Music Hack Day 2015, and won the incubation award from #MusicBricks. After 3 months of partial dedication, we created a prototype which I demoed in MTF Central 2015, MTF Berlin 2016, BLE Music 2016, and other events. More information in our website