Joan
Serrà
Profile
Joan did an MSc and PhD in machine learning for audio at the Music Technology Group of Universitat Pompeu Fabra (2006-2011) and a postdoc in artificial intelligence at IIIA-CSIC (2011-2015). After that, he joined Telefónica R&D as a machine learning researcher (2015-2019) and Dolby Laboratories as an AI researcher and research manager (2019-2024). He is currently with Sony AI, where he performs research on machine learning, focusing on audio and multimedia analysis, synthesis, and retrieval.
Publications
This paper explores the use of unlearning methods for training data attribution (TDA) in music generative models trained on large-scale datasets. TDA aims to identify which specific training data points contributed to the generation of a particular output from a specific mod…
Audio fingerprinting (AFP) allows the identification of unknown audio content by extracting compact representations, termed audio fingerprints, that are designed to remain robust against common audio degradations. Neural AFP methods often employ metric learning, where repres…
This paper explores the use of unlearning methods for training data attribution (TDA) in music generative models trained on large-scale datasets. TDA aims to identify which specific training data points contributed to the generation of a particular output from a specific mod…
We introduce the Robust Audio Watermarking Benchmark (RAW-Bench), a benchmark for evaluating deep learning-based audio watermarking methods with standardized and systematic comparisons. To simulate real-world usage, we introduce a comprehensive audio attack pipeline with var…
Detecting musical versions (different renditions of the same piece) is a challenging task with important applications. Because of the ground truth nature, existing approaches match musical versions at the track level (e.g., whole song). However, most applications require to …
JOIN US
Shape the Future of AI with Sony AI
We want to hear from those of you who have a strong desire
to shape the future of AI.
