Lightnews — Scholar-powered news

Prem Seetharaman @pseeth.bsky.social · Apr 2

Fun model to work on! More fun stuff to come!

Justin Salamon @justinsalamon.bsky.social · Apr 2

Generative Extend just released in Premiere Pro! Use GenAI to extend your video *and audio* clips for a perfectly timed edit.

The audio model was built by our team, the Sound Design AI (SODA) group at Adobe Research w/ @pseeth.bsky.social and @urinieto.bsky.social 🙌

www.youtube.com/watch?v=_Bv5...

Generative Extend | Premiere Pro 2025 Updates | Adobe Video

YouTube video by Adobe Video & Motion

www.youtube.com

1

Prem Seetharaman @pseeth.bsky.social · Dec 21

good story from someone completely unrelated to me i swear

Deepa Seetharaman @deepa.bsky.social · Dec 21

We’ve been hearing so much about reasoning lately but what happened to OpenAI’s big project: GPT-5 aka Orion?

It was to be a “significant leap forward” per Altman. Microsoft expected to see GPT-5 in mid-2024, per sources. But one problem after another popped up which…

www.wsj.com/tech/ai/open...

The Next Great Leap in AI Is Behind Schedule and Crazy Expensive

The startup has run into problem after problem on its new artificial-intelligence project, code-named Orion.

www.wsj.com

3

Prem Seetharaman @pseeth.bsky.social · Dec 14

👀

Jonathan Le Roux @jonathanleroux.bsky.social · Dec 14

@waspaa.com is on Bluesky!
Big changes are coming for WASPAA 2025!

IEEE WASPAA 2025 @waspaa.com · Dec 13

WASPAA is moving, for the first time in its almost 40-year history. Stay tuned for the announcement of the new venue!

3

Prem Seetharaman @pseeth.bsky.social · Dec 12

neat - i think all these spaces are basically a linear layer / permutation away from each other. with one codebook (or a vae setup) you could maybe just solve it with the embedding matrices directly, no audio needed

1

Prem Seetharaman @pseeth.bsky.social · Dec 12

Great work from @hugofloresgarcia.bsky.social’s internship at Adobe - turn your voice into basically anything!

hugofloresgarcía @hugofloresgarcia.bsky.social · Dec 12

new paper! 🗣️Sketch2Sound💥

Sketch2Sound can create sounds from sonic imitations (i.e., a vocal imitation or a reference sound) via interpretable, time-varying control signals.

paper: arxiv.org/abs/2412.08550
web: hugofloresgarcia.art/sketch2sound

1 3

Reposted by Prem Seetharaman

Kris Kashtanova @kris.art · Nov 27

Introducing MultiFoley, a video-aware audio generation method with multimodal controls! 🎉

⌨️Make a typewriter sound like a piano 🎹
🐱Make a cat meow like a lion roars! 🦁
⏱️Perfectly time existing SFX 💥 to a video

Link to research in comments:
by Adobe Research

2 5 41

Prem Seetharaman @pseeth.bsky.social · Nov 27

Check out our new work on video-guided audio gen with a focus on fine-grained creative control! Done by @czyang.bsky.social during an internship with our group at Adobe Research. Super fun model!

Ziyang Chen @czyang.bsky.social · Nov 27

🎥 Introducing MultiFoley, a video-aware audio generation method with multimodal controls! 🔊
We can
⌨️Make a typewriter sound like a piano 🎹
🐱Make a cat meow like a lion roars! 🦁
⏱️Perfectly time existing SFX 💥 to a video.

arXiv: arxiv.org/abs/2411.17698
website: ificl.github.io/MultiFoley/

2 10

Reposted by Prem Seetharaman

Simon Niklaus @sniklaus.com · Nov 26

A nifty application of depth estimation, creating a mockup of a digital design on real-world objects: sniklaus.com/mockup

2 9

Prem Seetharaman @pseeth.bsky.social · Nov 24

Here's one that seems to catch a bit more "thread-like" content, sorts by recency instead of likes, and drops arxiv bots: bsky.app/profile/psee.... Seems to work ok for now, and catches some non-ML threads too

Prem Seetharaman @pseeth.bsky.social · Nov 24

Made a feed that tries to index paper threads only: bsky.app/profile/psee.... To get into the feed, make a post with "arxiv.org" in the post somewhere + don't be a bot. My tiny contribution to the recent migration! Built w/ @skyfeed.app. Planning on some paper threads of my own soon...

1

Reposted by Prem Seetharaman

Prem Seetharaman @pseeth.bsky.social · Nov 24

Made a feed that tries to index paper threads only: bsky.app/profile/psee.... To get into the feed, make a post with "arxiv.org" in the post somewhere + don't be a bot. My tiny contribution to the recent migration! Built w/ @skyfeed.app. Planning on some paper threads of my own soon...

2 7

Reposted by Prem Seetharaman

Kashyap Chitta @kashyap7x.bsky.social · Nov 24

For those of you who haven't yet, give scholar-inbox.com a try! It's a free personal paper recommender which helps you stay up-to-date by sending daily/weekly paper digests directly to your inbox. Your votes train your own classifier, and you can have a peek at its feature words. Here are mine!

2 6 17

Prem Seetharaman @pseeth.bsky.social · Nov 24

Made a feed that tries to index paper threads only: bsky.app/profile/psee.... To get into the feed, make a post with "arxiv.org" in the post somewhere + don't be a bot. My tiny contribution to the recent migration! Built w/ @skyfeed.app. Planning on some paper threads of my own soon...

2 7

Reposted by Prem Seetharaman

Jonathan Le Roux @jonathanleroux.bsky.social · Nov 18

I initiated a starter pack for Audio ML. Let me know if you'd like to be added/removed.
go.bsky.app/LGmct4z

46 22 67