António Sá Pinto

[António Sá Pinto @asapintoPT]

@keunwoochoi @OrtetAI @kchonyc @HenriDwyer @elmanmansimov @causalclaudia Good luck, Keunwoo!

[António Sá Pinto @asapintoPT]

@kasaitakara Love this!! Where can I hear more of this?

[Keunwoo Choi @keunwoochoi]

the Kimi people @Kimi_Moonshot successfully prompted me to prompt these 17 LLMs to make these 17 Beetboxes 🥁🪩 who's grooving the best?

play more: https://t.co/zpgCaP7xgt

[zilla 🌱 @hey_zilla]

ok i think suffice to say that Opus 5 is pretty good at Python... sound on ;)

[António Sá Pinto @asapintoPT]

@hey_zilla Wow!!🤩

[António Sá Pinto @asapintoPT]

@joelhooks How do you show that glyph after each agent?

[António Sá Pinto @asapintoPT]

@made_by_dy Curious about this…are y gonna share more info?

[António Sá Pinto @asapintoPT]

@dwhitedesign Still can’t believe its only 2k!! Really nice

[Mirelo @MireloAI]

Today, together with @kyutai_labs, we’re introducing our new Audio-to-MIDI model.

It takes a finished recording, identifies the instruments playing, and returns separate MIDI tracks for each — voice, drums, bass, keys, and more.

Unlike most existing solutions, our model works directly from the full mix rather than requiring separate stems.

It also detects chords, key, and tempo, giving producers broader musical context.

We’ve written more about the model, the problem, and how it works here: https://t.co/GvzH7hLjlg

[Chris Donahue @chrisdonahuey]

Vibe coding is cool but have you tried vibe patching?

Pure Vibes = Pure Data + MCP. Describe your sound, watch the patch appear 🌊

🔊👇

[António Sá Pinto @asapintoPT]

@chrisdonahuey This is so cool! Thanks for sharing

[ISMIR Conference @ISMIRConf]

And the winner of the ISMIR 2025 10-Year Test-of-Time Award is... a true landmark in rhythmic analysis! ✨🏆

A huge, heartfelt congratulations to Florian Krebs, Sebastian Böck, and Gerhard Widmer for their paper, "An Efficient State-Space Method for Joint Tempo and Meter Tracking."

This work provided a brilliant and elegant "drop-in replacement" for beat and meter tracking systems that achieved a rare triple-win: it was more accurate, drastically faster, and more memory-efficient. A decade later, its influence is still felt in the robust and practical tools our community relies on. This paper set a new standard for a fundamental MIR task. A truly deserving winner!

#ISMIR2025 #TestOfTimeAward #BeatTracking #MIR #SignalProcessing #LandmarkPaper

[ISMIR Conference @ISMIRConf]

Honoring a true classic from the archives! The #ISMIR2025 Classical Test-of-Time Award goes to Meinard Müller et al. for their 2005 paper, "Audio Matching via Chroma-Based Statistical Features." 🎼

This paper introduced a robust chroma feature that revolutionized audio matching for classical music. Two decades later, it remains a pillar of the field. Truly foundational work. Congratulations!

#TestOfTime #MIR #ChromaFeatures #ClassicalMusic

[ISMIR Conference @ISMIRConf]

A field stands on two legs: algorithms & data. Sharing the #ISMIR2025 Classical Test-of-Time Award is the legendary Masataka Goto for his 2006 AIST Annotations for the RWC Database.📜

This work provided THE benchmark dataset for a generation of MIR research. A monumental contribution that enabled progress across the field. A true honor to celebrate this work tonight!

#TestOfTime #MIR #OpenData #GroundTruth

[António Sá Pinto @asapintoPT]

@keunwoochoi It was perfect!

[António Sá Pinto @asapintoPT]

@keunwoochoi Thanks for sharing. Super helpful for a first-timer! Also, that 풍월당 Pungwoldang got me really curious :)

[Chris Donahue @chrisdonahuey]

Excited to share our beta release of Music Arena, a live evaluation platform for state-of-the-art AI music generation models!

🎧 Listen to the latest models and 🗳️ vote for your favorite

⚔️ https://t.co/XyXvzlOMcH
⭐️ https://t.co/9f3mvseEyu
📜 https://t.co/HqTg5PH09O

[Keunwoo Choi @keunwoochoi]

please book your accommodation at #ISMIR2025 either i) directly through links at https://t.co/Jd4T6kNZRb or ii) just DIY.

https://t.co/523bgyqnbj is not affiliated with us, it is a well-known scam.
https://t.co/jW2sgBcNGM

[Pablo Alonso Jiménez @pablo__alonso]

Introducing OMAR-RQ – a new family of self-supervised audio encoders for music understanding!

Trained on 330K+ hours of music using masked token prediction.

🔗 Paper: https://t.co/xXZxLiMHxD
🔗 Models: https://t.co/4dXKMHDBZv

[António Sá Pinto @asapintoPT]

Our @ISMIRConf 2025 paper is now on arXiv!
We explore onset detection in low-data Maracatu 🥁 via transfer learning: intra & cross-task (beat→onset), layer-wise fine-tuning. F1↑ up to 50p.p.
📄 https://t.co/vZXCluiGSh
💻 https://t.co/hEieyc6Utv
Let’s connect at #ismir2025 🇰🇷