MTG researchers participated to the 45th International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2020 (https://2020.ieeeicassp.org/). Originally it had to take place in Barcelona from May 4th to the 8th 2020 but because of the Covid-19 pandemic it finally happen virtually. The IEEE Signal Processing Society (SPS) offered ICASSP attendance as complimentary, so anyone will be able to participate.

Here are the articles from MTG researchers that were presented:

• Pablo Alonso-Jiménez, Dmitry Bogdanov, Jordi Pons, Xavier Serra. "Tensorflow Audio Models in Essentia."

Publication: https://ieeexplore.ieee.org/document/9054688/

Video: https://youtu.be/25YPvwIC-aE

• Minz Won, Sanghyuk Chun, Oriol Nieto, Xavier Serra. "Data-Driven Harmonic Filters for Audio Representation Learning."

Publication: https://ieeexplore.ieee.org/document/9053669/

Video: https://youtu.be/BXHEYb_Axus

• Pritish Chandna, Merlijn Blaauw, Jordi Bonada, Emilia Gomez: "Content Based Singing Voice Extraction From a Musical Mixture."

Publication: https://ieeexplore.ieee.org/document/9053024/

Video: https://youtu.be/H0vryP4jn1k

• Merlijn Blaauw, Jordi Bonada: "Sequence-to-sequence Singing Synthesis Using the Feed-forward Transformer."

Publication: https://ieeexplore.ieee.org/document/9053944/

Video: https://youtu.be/1EXnFmwgJNw

• Jordi Bonada, Merlijn Blaauw: "Hybrid Neural-Parametric F0 Model for Singing Synthesis."

Publication: https://ieeexplore.ieee.org/document/9054707/

Video: https://youtu.be/v2PQGydDf-o

• António Ramires, Pritish Chandna, Xavier Favory, Emilia Gómez, Xavier Serra. "Neural Percussive Synthesis Parameterised by High-Level Timbral Features."

Publication: https://ieeexplore.ieee.org/document/9053128/

Video: https://youtu.be/NeB-jTI9GzI

• Furkan Yesiler, Joan Serrà, Emilia Gómez. "Accurate and Scalable Version Identification Using Musically-Motivated Embeddings."

Publication: https://ieeexplore.ieee.org/document/9053793 

Video: https://youtu.be/8VuKRaFntRE

• Álvaro Sarasúa, Jordi Janer, Oscar Mayor, Jordi Bonada, Merlijn Blaauw. "Choir Singing Synthesis For Rehearsal Tools With Large-Scale Multilingual Repertoires"

Video: https://youtu.be/9vEA4iSjajg

Xavier Serra gave an Eton talk on "Signal Processing and Machine Learning Challenges in Sound and Music Computing". https://2020.ieeeicassp.org/program/eton-talks/