Happy to share our new paper: Real-time Timbre Remapping with Differentiable DSP Timbral control of synthesizers (like an 808) in real-time using audio from acoustic percussion performances. 🥁🎛️ arXiv: arxiv.org/abs/2407.04547 audio/video/code: jordieshier.com/projects/nime2…

🚨 Call for Papers: First AES International Conference on Artificial Intelligence and Machine Learning for Audio (AIMLA 2025), Queen Mary University of London, Sept. 8-10, 2025. c4dm.eecs.qmul.ac.uk/news/2024-07-2…

We're happy to share that our students Soumya Sai Vanka Franco Caspe and Farida Yusuf are part of the organizing committee for the AIMLA conference next year. We're also calling for contributions; please check the blog for details. aim.qmul.ac.uk/cfp-first-aes-…

Our student Jiawen will present "Towards Building an End-to-End Multilingual Automatic Lyrics Transcription Model," an extension of her UKIS poster, at EUSIPCO 2024 at the end of August. aim.qmul.ac.uk/aim-at-eusipco…

🔊 Encode and decode audio to/from latents with Music2Latent! 🔊 Music2Latent encodes only ~10 latents per second of audio 👀 This means lightning-fast training/inference of latent generative models ⚡️️ Try it: github.com/SonyCSLParis/m… Paper: arxiv.org/abs/2408.06500 How? 👇🧵

Send in your ideas for hosting challenges at AES AIMLA 2025

Differentiable Time-Varying Linear Prediction in the Context of End-to-End Analysis-by-Synthesis Paper: arxiv.org/abs/2406.05128 Web: yoyololicon.github.io/golf2-demo/ Improvements to the GOLF voice synthesiser and comprehensive comparison. Details in this thread. 1/7

📢📢Glad to share our latest survey on Foundation Models for Music! 🎶For both computer music researchers and ML researchers: this comprehensive review explores how LLMs and diffusion models are revolutionising music through music understanding, generation, therapy, and more.🎉🎉

AFX-Research: an Extensive and Flexible Repository of Research about Audio Effects All the research on audio effects of the last few decades. A table with lots of metadata. Search. Filter. Order. Contribute. repo: github.com/mcomunita/AFX-… web: mcomunita.github.io/AFX-Research

NIME is happening next week! This time, our members Jordie Shier and Shuoyang will present two papers on DDSP and interactive ML, respectively, and teresa pelinski will lead two workshops on embedded AI. Details: aim.qmul.ac.uk/aim-at-nime-20…

**Correction** Teresa is leading workshops on " First—and second-person perspectives for ML in NIME" and "Building NIMEs with Embedded AI." In the previous tweet, we mistakenly said they were about the same topic.

Just out! New ACM TOCHI paper with Courtney Reed Adán L. Benito Franco Caspe: "Shifting Ambiguity, Collapsing Indeterminacy: Designing with Data as Baradian Apparatus". Open access: dl.acm.org/doi/10.1145/36…

Next week, we will host a seminar on "Machine Learning-Based Artificial Reverberation & Non-stationary Noise Removal from Repeated Sweep Measurements", by Gloria Dal Santo & Sebastian J. Schlecht, respectively. More information at: c4dm.eecs.qmul.ac.uk/news/2024-09-1…

Yesterday Marco Martínez and I presented a hands-on tutorial on AI for Multitrack Mixing at @dafx2024 at the University of Surrey. Find our slides: drive.google.com/file/d/1-lS_KN…

Marco Martínez University of Surrey and finally an up-to-date list of research in the field at csteinmetz1.github.io/AutomaticMixin…

Mid-July I presented my work as part of a case study on responsible and ethical design of AI systems organised by the UAL Creative Computing Institute as part of their MusicRAI research project. Slides at music-rai.github.io

aes2.org/contributions/… AES AIMLA 2025 website is now up to date with submission deadlines. The submission portal for challenge and tutorial proposals is now open until the end of Oct 2024. #AESAIMLA25

"Differentiable All-pole Filters for Time-varying Audio Systems" being presented by Chin-Yun Yu at DAFx! Paper, audio samples, code, and plugins available at diffapf.github.io/web/