Deezer Research

Research team @Deezer - Understanding music and how people relate to it #Music #MIR #RecSys #MachineLearning - also on @[email protected]

[Guillaume Salha-Galvan @gsalhagalvan]

Exciting news 🥳🍾
I am honored to have been shortlisted, for the third time in my career, among the best paper candidates at the @ACMRecSys conference!

After nominations in 2020 and 2021, this time it’s at #RecSys2024 with Viet-Anh Tran, @bmsguerra, and @romainhennequi1!

[1/3]

[Guillaume Salha-Galvan @gsalhagalvan]

A nice way to start the week! Our work:

"Transformers Meet ACT-R: Repeat-Aware and Sequential Listening Session Recommendation"
led by Viet Anh Tran from @Deezer,

has been accepted as a Full Paper at #RecSys2024! 🥳

We will share a preprint soon along with code and data.

[Deezer Research @researchdeezer]

STONE : Self-supervised Tonality Estimator by Yuexuan Kong, Vincent Lostanlen Gabriel Meseguer Brocal Stella Wong, Mathieu Lagrange and Romain Hennequin https://t.co/23YpiUEX88 STONE is the first self-supervised tonality estimator using the ChromaNet architecture.

[Deezer Research @researchdeezer]

Paper news! Two submissions by #deezer team members have been accepted at @ISMIRConf :

[Deezer Research @researchdeezer]

From Real to Cloned Singer Identification by Dorian Desblancs, Gabriel Meseguer Brocal, Romain Hennequin and Manuel Moussallam: https://t.co/RrU4V3eBOp where we investigate ways to identify artist's cloned voices in mixtures.

[Deezer Research @researchdeezer]

Our Head of Research Romain Hennequin will be at @MusicTechBE to talk about #AImusic and the latests work from the team on the topic. https://t.co/kBI0AVQoh8

[Deezer Research @researchdeezer]

The deadline to submit your research to the NLP4MusA #ISMIR2024 companion workshop has been extended to July 12th: https://t.co/TOSEz9XT2r . Both academic and industry submissions at the crossroads of #NLP and #MIR are welcomed

[Guillaume Salha-Galvan @gsalhagalvan]

We will present vMF-exp for scalable exploration in RL problems with large action sets and hyperspherical embeddings, next month at the @arlet_workshop of #ICML2024!

Paper online soon: a math-heavy work with a real-world deployment on @Deezer for music recommendation! 🎶

(1/2)

[Jordi Pons @jordiponsdotme]

New post is out! 🇰🇷
ICASSP 2024, eleven picks.
https://t.co/0BJZzfjU1J

[Deezer Research @researchdeezer]

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

[Guillaume Salha-Galvan @gsalhagalvan]

I'm thrilled and humbled to share that our work on new releases has just been awarded the "Best Industry Track Paper Award" at #ECIR2024!

Congratulations to all members of our music recommendation team at @Deezer! 👏

@ecir2024 @researchdeezer @lea_ibrd @xbenji

[Guillaume Salha-Galvan @gsalhagalvan]

This week, Glasgow plays host to the #ECIR2024 conference! 🏴󠁧󠁢󠁳󠁣󠁴󠁿

@lea_ibrd from our team is set to present our initiatives to improve the exposure and recommendation of new music releases on @Deezer! 🎶

Check it out:
Paper: https://t.co/A4FIXYlz16
Slides: https://t.co/z8PIXiE9vI

[Deezer Research @researchdeezer]

We are represented at #CHIIR24 with Lilian Marey presenting his work on modeling music consumption patterns https://t.co/0S6kG6csMx done at #Deezer with @bmsguerra and Manuel Moussallam

[Deezer Research @researchdeezer]

News from our #NLP team! @gaspmichel PhD student in our team present his last work done with @evpure, @romainhennequi1 and @ccerisara at the LaTeCH-CLfL 2024 SIGHUM Workshop in Malta. Preprint available here: https://t.co/me7eWEEzh2 #NLP

[Deezer Research @researchdeezer]

Congratulations to Kristina Matrosova, Manuel Moussallam, Thomas Louail, and Olivier Bodini for their paper "Depict or Discern? Fingerprinting Musical Taste from Explicit Preferences" published in @TismirJ! 👏

Paper link: https://t.co/xQgLwmpkjX

[TISMIR journal @TismirJ]

NEW: Research paper by Kristina Matrosova @MMoussallam @ThomasLouail and Olivier Bodini: "Depict or Discern? Fingerprinting Musical Taste from Explicit Preferences" shows with a dataset that the 2 opposite views about taste lead to conflicting solutions
https://t.co/XHsPl98bJT

[Deezer Research @researchdeezer]

Join us tomorrow for this exciting #RecSys MeetUp, co-organized by @Deezer! We're gathering at @leboncoin's Paris office with speakers from @dailymotion, @CarrefourFrance, and @leboncointech.

More details here: https://t.co/YUnkRn7Piv

@MeetupFR @recsysfr @DeezerDevs @grillatg

[Deezer Research @researchdeezer]

More good news! Our paper "An Experimental Comparison of Multiview Self-supervised Methods For Music Tagging" has recently been accepted at #ICASSP2024.🎶🇰🇷
Kudos to @Gabolsgabs, Dorian Desblancs, and @romainhennequi1!

[Gabriel Meseguer @Gabolsgabs]

Our paper "An Experimental Comparison of Multiview Self-supervised Methods For Music Tagging" was accepted at #ICASSP2024!
Work with Dorian Desblancs and @romainhennequi1 which compares different self-supervised methods for music tagging.
Code and preprint soon. Stay tuned!

[Deezer Research @researchdeezer]

Congratulations! 👏This #ECIR2024 paper is now available online: https://t.co/Hjo4QP9wGZ

@ecir2024 @lea_ibrd @gsalhagalvan

[Guillaume Salha-Galvan @gsalhagalvan]

Exciting news to end the year... 🤩

Our paper on improving the discoverability of new releases on @Deezer is set for an oral presentation at the Industry Day of #ECIR2024 in Glasgow!

Kudos to @lea_ibrd and the entire @Deezer RecSys team! 👏 The paper will be online soon.