Genís Plaja

We have just released a new version of compIAM :) Check it out here: github.com/MTG/compIAM and pip install compiam

📢 New preprint version of "Transparency in Music-Generative AI: A Systematic Literature Review" is out! We now analyse 107 publications in a two-stage review, uncovering gaps in transparency for genAI in music. Jump to section 5.2 for TL;DR ✨ doi.org/10.21203/rs.3.…

tomorrow i’ll be presenting at @ISMIRConf for the first time! 💫 i’ll be sharing our work on data replication assessment in AI gen music and introducing MiRA 👀 #ismir2024

We are more than open to contributions! Feel free to get in touch if you want to make this initiative to grow bigger. We've also updated to documentation with all the improvements for v0.4.0: mtg.github.io/compIAM/index.…

We will present some of the tools and datasets in compIAM today at the ISMIR Conf. within the context of our TISMIR work: Repertoire-Specific Vocal Pitch Data Generation for Improved Melodic Analysis of Carnatic Music. Big thanks to all co-authors❤️

We have just released a new version of compIAM :) Check it out here: github.com/MTG/compIAM and pip install compiam

it's time to pip install mirdata :) thanks everyone for the great work!!

📢 Call to all #MIRCommunity! Have you ever wondered what an open model means? Help us shape the definition of open models in generative AI for music by taking our survey — just 10 minutes! 👉 forms.gle/Z48t6HPBXwWC3r… thank you 💫

happy to announce that MiRA is out!💫 MiRA (Music Replication Assessment) is an open model-independent tool to assess audio data replication of the training set 🎶🔧🤖👀 installable as a PyPI package 🎁 github.com/roserbatlleroc…

We are happy to announce the publication of our latest paper in @JOSS_TheOJ presenting Soundata, a library for working with audio datasets in a standarized way + today's presentation by @genisplaja of Soundata at @ecoacoustics24 congress! upf.edu/web/mtg/home/-…

We’re happy to announce our recent JOSS paper! 🎉Soundata: Reproducible use of audio datasets ⚙️github.com/soundata/sound… 📄 joss.theoj.org/papers/10.2110…

Special Collection "Cultural Diversity in MIR Research". The Editorial by Zhiyao Duan, @pvankranenburg, @juhan_nam and Preeti Rao has just been published. Continues the effort initiated in ISMIR 2021. transactions.ismir.net/articles/10.53…

@danperalbo14 fota-li Dan :) :)

Recordo aquest tuit d'en Pau de fa un temps! Em va quedar ben clavat! Doncs avui sóc jo qui ho ha fet! I sí, és un dia molt important! Esperem que hi hagi suficient aigua a la piscina perquè m'hi tiro de cap! Amb por i emoció, pro m'hi tiro! 🚀 Lesgereee

@roserbatlleroca @thomasgnuttall @MorsiAlia @ETIC_UPF @casalinfants @youngitgirls @ruiz_aurelio dream team!!

divendres amb en @thomasgnuttall i la @MorsiAlia vam realitzar un taller sobre Música i Entorn, per la Jornada Tictaker organitzada entre @ETIC_UPF @casalinfants @youngitgirls, i va ser molt xulo ✨✨ gràcies a l’@ruiz_aurelio per motivar-nos a col·laborar en projectes així 😊

We are MTG (@mtg_upf) 🎶 See you next year in San Francisco 🌉@ISMIRConf

@nkundiushuti Marius🫶🏽 Thanks for your advise and support!! It's always a pleasure to do research with you :)

your source separation training data is noisy (bleeding between sources)? turns out you can still train your (cold diffusion) model and then use clustering on all the diffusion steps to clean the source signal. great ISMIR paper by the source separation jedi, @genisplaja

Thanks to all amazing people involved in making that possible!!

compiam is open to contributions, new datasets, and models. Thanks to mirdata v0.3.8 we do now have easy and standardized access to 10 Indian Art Music datasets. Let’s keep it growing :)

Since we’ve had really cool software releases and upgrades during @ISMIRConf this year, why not having one more? :) We just released compiam v0.3.0, including multiple improvements and new features!