Time for score image-audio alignment!!
🎶How can we learn to follow along to a music performance in a score without direct supervision?🎶
Introducing FuSiLi (Fused Sinkhorn-Localized Similarity)!
A multimodal contrastive learning approach that learns precise local alignments using only global supervision.
🧵
11
1,433
If you are attending #ismir2025, do not miss the Tuesday evening event!🇰🇷 We invited one of the best Korean traditional music ensembles, with awesome K-food, drink, and entertaiment(🔴🔺🟥 )+ chance to try Korean instruments🤩 More info is coming soon!⌛️
This score-image→audio generatioin model was trained only with < 20 seconds segments, but Jongmin somehow implemented it to contuning the generation through entire piece!
Dasaem Jeong retweeted
🔥Happy to announce that the AI for Music Workshop is coming to #NeurIPS2025!
We have an amazing lineup of speakers! We call for papers & demos (due on August 22)!
See you in San Diego!🏖️
@chrisdonahuey @Ilaria__Manco @zawazaw @huangcza @McAuleyLabUCSD @zacknovack @NeurIPSConf
Time to register ISMIR 2025!
📢 Registration is now OPEN for #ISMIR2025!
Join us in Daejeon, Korea 🇰🇷 from Sep 21–25, 2025
for the 26th International Society for Music Information Retrieval Conference.
Due to the delay in opening the registration, we have extended the deadlines.
🗓️ Updated Key Deadlines (AOE):
➡️ Author Registration: Jul 7 → Jul 18
➡️ Early Registration: Jul 28 → Aug 4
➡️ Regular Registration: remains Sep 10
🔗 Register here:
ismir2025.ismir.net/registra…
12
1,094
Dasaem Jeong retweeted
🏨 Attending #ISMIR2025 (Sep 21–25 in Daejeon)?
‼️ Multiple conferences are happening that week —
👉 Hotels are expected to fill up fast.
⚠️ Don’t wait — book your accommodation as soon as possible!
🔗 Info & options:
ismir2025.ismir.net/accommod…
#MIR #MusicTech #MusicAI
Excited to announce this superb list of tutorials!!
🎓 ISMIR 2025 Tutorials – Sunday, Sep 21! 🎓
Check out exciting tutorials at the link below!
🔗 ismir2025.ismir.net/program-…
#ISMIR2025 #MIR #MusicTech #Daejeon #Korea
There's a new version of this post
4
136
Dasaem Jeong retweeted
📣 Save the Date!📣
The 26th International Society for Music Information Retrieval Conference is coming this September 21st-25th in Daejeon, Republic of Korea! 🇰🇷
For more information, visit ismir2025.ismir.net
#ISMIR2025 #Daejeon #Korea
🎶Now a neural network can read scanned score image and generate performance audio in end-to-end😎
I'm super excited to introduce our work on Unified Cross-modal translation between Score Image, Symbolic Music, and Audio.
Why does it matter and how to make it? Check the thread🧵
By training a model to generate audio tokens from given score image, the model learn how to read notes from the score image. This led our model to break SOTA for OMR! Vice versa for AMT can work, while the gain was not significant enough compared to the OMR.
1
7
665
You can refer details below:
arXiv: arxiv.org/abs/2505.12863
demo: sakem.in/u-must/
Work done by Jongmin Jung(@sake_min), DongMin Kim, Sihun Lee(@is_s_yun), Seola Cho, Hyungjoon Soh, Irmak Bukey, Chris Donahue(@chrisdonahuey) and me.
Dasaem Jeong retweeted
hi all, hi audio gen folks.
please check out KAD! we truly believe KAD is a great alternative to FAD. why? thread:
``KAD: No More FAD! An Effective and Efficient Evaluation Metric for Audio Generation,'' Yoonjin Chung, Pilsun Eu, Junwon Lee, Keunwoo Choi, Juhan Nam, Ben Sangbae Chon, ift.tt/HYxTXkB
Dasaem Jeong retweeted
PESTO 2.0 è rilasciato! 🥳🥳🥳
With Brazilian chef @torres_be_ (and others), we revisit this traditional italian sauce, invented in Milan at @ISMIRConf 2023 🇮🇹
And you can taste it in REAL-TIME at home (~5 ms latency) ⏱️
1/6
Dasaem Jeong retweeted
🚀 We’re excited to introduce CLaMP 3: A breakthrough in multimodal & multilingual music retrieval! 🎵🔍
It unifies audio, sheet music, MIDI, and text across 100 languages into a shared space, enabling cross-modal retrieval & emergent alignment with unseen languages.
🧵[1/5]👇
Our paper on 🎻 synthesis got accepted to ICASSP! (with @hermanhwdong )
We used a dataset transcribed by Nazif’s last ISMIR paper that includes pitch bend info. We explicitly modeled these pitch bend for better performance!
📝: arxiv.org/abs/2409.12477
🎶: daewoung.github.io/ViolinDif…
서강대 퓨처랩 연구소에서 "AI와 VR을 활용한 몰입형 음악 교육 콘텐츠 개발 및 적용" 과제 수행을 위한 포닥 전임연구원을 채용합니다. AI, VR, 디지털 미디어, HCI, 음악 교육 등 다양한 분야의 박사님들을 초빙하니 많은 관심 부탁드립니다.
creative.sogang.ac.kr/%ed%8f…
















