This paper I contributed to is accepted to LREC-Coling 2024 LREC COLING 2024 !

I love music most when it’s live, in the moment, and expressing something personal. This is why I’m psyched about the new “DJ mode” we developed for MusicFX: aitestkitchen.withgoogle.com/tools/music-fx… It’s an infinite AI jam that you control 🎛️. Try mixing your unique 🌀 of instruments, genres,

I have collaborated with my tattooist friend, j.ta.2, on AI x Tattoo. This collaboration has been featured in an interview with Visla Magazine. 근 1년간 타투이스트 친구, j.ta.2와 진행한 협업, AI x 타투에 관련해 비즐라 매거진에서 인터뷰를 진행하였습니다. visla.kr/feature/250668/

With text-to-music models really starting to heat up, today I'm sharing a new, open model 🚀 🎶 musicgen-songstarter-v0.2 🎶 A large, stereo MusicGen that acts as a useful tool for music producers Compared to v0.1, its 2x the size & trained on 3x more hand-curated samples

A new paper I participated ‘Nested Music Transformer’ got accepted for ISMIR Conference !

Self-Attention by hand ✍️ Excel ~ I designed this exercise for students to practice the QKV math. I also created a medium and a large version to show how the attention matrix grows quadratically as the sequence gets longer. 👇Join the 'AI Math' community. Download xlsx.

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%ac%eb%8…

Successfully finished presenting "MusicGen-Chord" at the LBD section of ISMIR Conference. Thanks for the support Andreas Jansson and Dasaem Jeong! #ISMIR2024

✨ Train language models directly on continuous data - without tokenization ✨ We propose an easy way to train GPT-style autoregressive models on continuous data, without error accumulation. We test it on audio 🔊, but this method can easily work with other modalities 🎆 👇🧵

Sorting Algorithm Score Video 1: Selection Sort

A new audio-symbolic-text joint embedding has been released (Like ImageBind!). Use it for music retrieval with multilingual queries, Conditioning, RAG, FD Score, and more!

Two undergrads. One still in the military. Zero funding. One ridiculous goal: build a TTS model that rivals NotebookLM Podcast, ElevenLabs Studio, and Sesame CSM. Somehow… we pulled it off. Here’s how 👇

My recent work "Unified Cross-modal Translation of Score Images, Symbolic Music, and Performance Audio" is now on arxiv! arxiv.org/abs/2505.12863 Demo: sakem.in/u-must