👇👇👇I'll be presenting our XAI-SA workshop paper today at 14:00 KST. If you are at #ICASSP24 come say hi👋

We are at #ICASSP2024 to present a paper by Gabriel Meseguer , Dorian Desblancs and Romain Hennequin on Self-supervized methods for Music Tagging

Presentation and open discussion with Moisés Horta Valenzuela (𝔥𝔢𝔵𝔬𝔯𝔠𝔦𝔰𝔪𝔬𝔰) + roser batlle roca and bkotowski on critical art practices within Generative Music AI. Join the event at UPF Barcelona EnginyEria - UPF or online. 🗓️April 29 at 6:30PM. Info 👉 upf.edu/web/mtg/home/-…

Shuo Zhang, Senior Machine Learning Research Engineer at Bose Research, will offer the seminar "Better audio through AI research at Bose" at the MTG EnginyEria - UPF. Friday, May 3rd at 12:00h: upf.edu/web/mtg/home/-…

NLP4MusA is back! Happy to share that I'm co-organising the 3rd Workshop on NLP for Music and Audio co-located with ISMIR Conference. We’ll be opening submissions soon, but in the meantime CfP and more info here: sites.google.com/view/nlp4musa-…

Freesound's take on the usage of its database for the development of generative AI models: potential threats, opportunities, licensing issues, and more👇

Software company unilaterally grants itself the right to use and republish your work, and presumably to also vacuum it into their 'AI' plagiarism machine so that it can be sold back to you. #Enshitification continues apace (h/t Cory Doctorow NONCONSENSUAL BLUE TICK)

Thinking of organizing a Barcelona AI Music Meetup. Who's in?

We are happy to announce the publication of our latest paper in JOSS presenting Soundata, a library for working with audio datasets in a standarized way + today's presentation by Genís Plaja of Soundata at 5th World Ecoacoustics Congress - Madrid 2024 congress! upf.edu/web/mtg/home/-…

I am glad to announce that the work we carried out last summer at @sonycsl with Taketo (Taketo Akama) has been accepted to #ISMIR2024! 🎶🔍 Music Proofreading with RefinPaint: Where and How to Modify Compositions Given Context 📚 arxiv.org/pdf/2407.09099

Next week Pablo Alonso Jiménez defends his PhD thesis: "Deep Audio Representation Learning for Music Using Weak Supervision". 🗓️Thursday October 3rd, 2024 at 11:00h 📍UPF Campus Poblenou EnginyEria - UPF upf.edu/web/mtg/home/-…

🧑‍🎓Last week I defended my PhD thesis, Deep Audio Representation Learning for Music Using Weak Supervision at MTG-UPF. ❤️Huge thanks to my supervisors Dmitry Bogdanov and Xavier Serra, the jury, and all my friends and family for attending. Video of the defense available soon!

A joint embedding model that leverages a fine-tuned LLM to handle music descriptions from various datasets has been released. In addition to semantic queries (genre, instruments, mood, theme), it also supports metadata similarity queries for music search."

We have two new doctors at the MTG: Nazif Can Tamer and Pablo Alonso Jiménez. 🎓Congratulations!!!🎓The videos of their thesis defenses are now available here: youtube.com/playlist?list=…