Jonah Casebeer
@CasebeerJonah
Followers
197
Following
381
Statuses
38
Research Scientist @AdobeResearch | Ph.D. @IllinoisCS | personal account
Joined April 2020
RT @_akhaliq: Presto! Distilling Steps and Layers for Accelerating Music Generation Despite advances in diffusion-based text-to-music (TT…
0
29
0
Accelerate your diffusion model via step & layer distillation. Check it out applied to music! Work lead by the incredible @zacknovack while at @AdobeResearch. Thanks to @__gzhu__, @NicholasJBryan, @McAuleyLabUCSD, and @BergKirkpatrick.
Ultra-fast text-to-music generation w/o degrading quality? Introducing Presto! Distilling Steps and Layers for Accelerating Music Generation 🎹: 📖: w/@__gzhu__ @CasebeerJonah @BergKirkpatrick @McAuleyLabUCSD @NicholasJBryan 🧵
0
1
20
RT @ArxivSound: ``DITTO-2: Distilled Diffusion Inference-Time T-Optimization for Music Generation,'' Zachary Novack, Julian McAuley, Taylor…
0
6
0
RT @minje_research: Two big announcements! (a) We appreciate many inquiries about HSCMA 2024. We are happy to announce that the paper subm…
0
1
0
Very excited to be helping with this! We have two paper tracks, and an exciting speaker lineup. More info:
We organize HSCMA 2024 this year as a satellite workshop of ICASSP 2024. And we are calling for paper contributions! If you need an excuse to stay for another day in Korea during the ICASSP 2024 week, this will serve the purpose. 😉 Link to the CfP:
0
0
5
@ArxivSound Demo, code, and trained weights all linked in our thread.
Higher Order Meta-AF — Control adaptive filters with neural networks using inter-frequency information. Applied to echo cancellation & speech enhancement. Our work “Meta-Learning for Adaptive Filters with Higher-Order Frequency Dependencies” is up! See
0
0
1
Higher Order Meta-AF is a simple way to leverage frequency dependencies to simultaneously improve performance and reduce complexity. For an introduction to meta-learning for adaptive filters, check out our previous work! 6/N
Meta-AF — Control adaptive filters with neural networks. Applied to system identification, echo cancellation, equalization, dereverberation, and beamforming. Our pre-print: “Meta-AF: Meta-Learning for Adaptive Filters” is up! w/@NicholasJBryan & P. Smaragdis 1/N
0
0
1
RT @csteinmetz1: DeepAFx-ST: Automatic control of audio effects for transferring the production style from one recording to another using d…
0
62
0
We’re very excited to apply Meta-AF to other domains and tasks! Let’s discuss! 12/N, N=12
0
0
5