Prem Seetharaman

Researcher in computer audition, machine learning, and HCI. Sr. Research Scientist, @AdobeResearch. Previously @DescriptApp, @Northwestern.

[Nicholas J. Bryan @NicholasJBryan]

Introducing "DRAGON: Distributional Rewards Optimize Diffusion Generative Models"!

📖: https://t.co/biSceK5vgQ
🎹: https://t.co/6taiaC8SUZ

A new framework for fine-tuning gen models towards a target distribution.

By Yatong Bai w/@CasebeerJonah @somayeh_sojoudi @NicholasJBryan

[arXiv Sound @ArxivSound]

``DRAGON: Distributional Rewards Optimize Diffusion Generative Models,'' Yatong Bai, Jonah Casebeer, Somayeh Sojoudi, Nicholas J. Bryan, https://t.co/K5SvAHZ0YC

[Justin Salamon @justin_salamon]

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 @AdobeResearch w/ @pseetharaman and @urinieto 🙌

https://t.co/c6v56oxAhM

[Deepa Seetharaman @dseetharaman]

NEW story w/ @cityofthetown: chatter about o3 dominated X today. But it’s worth thinking abt OpenAI’s other big project: GPT-5 aka Orion, which has faced a slew of unexpected issues since mid-2023 that put it behind schedule.

MSFT expected to see GPT-5 about 6 months ago…

[Dreaming Tulpa 🥓👑 @dreamingtulpa]

adobe is cooking a new method to generate sound from vocal imitations 👀

[Prem Seetharaman @pseetharaman]

Cool work from our group at Adobe, led by our awesome intern @hugggof! Turn your voice (or anything) into anything!

[Deepa Seetharaman @dseetharaman]

I'm in Vancouver for NeurIPS until 12/13. Would love to chat about what’s coming in AI.

Find me on Signal (dseetharaman.23) if you’d like to chat.

[Justin Salamon @justin_salamon]

📢 Audio AI Job opportunity at Adobe!

The Sound Design AI Group (SODA) is looking for an exceptional research engineer to join us in building the future of AI-assisted audio and video creation.

Strong ML background, GenAI experience a plus.

Details: https://t.co/IOlA7Rlwhg

[Prem Seetharaman @pseetharaman]

New work from our group! Also, find this thread on the other place: https://t.co/T4eFQYGjwN :)

[Ziyang Chen @CzyangChen]

🎥 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

[Ziyang Chen @CzyangChen]

🎥 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

[New Lantern @newlanternai]

Today, we're emerging from stealth with our $19M Series A fundraise led by @ericvishria and @benchmark. We launched the first AI radiology resident.

The round is accompanied by @Afore @svangel @Neo Anthology Fund (@MenloVentures + @AnthropicAI) @jaykreps @rauchg @amasad @varsharao @pirroh Saji (Benchling), Jay Desai (Bamboo Health), Allison Pickens, Sabrina Hahn, and other amazing angel investors!

https://t.co/3h02NlMMZO

https://t.co/ifZulxkwvI

[Shiva Suri @shivasuri_]

Super excited to announce our $19M Series A led by @ericvishria / @benchmark!!

Radiologists play Where's Waldo? with medical images on super-hard mode all day. We're building the first AI Radiology Resident so radiologists work faster and easier.

https://t.co/Gj0WM5dlCs

[Shiva Suri @shivasuri_]

shameless self-promo; we're hiring pls help us get more stuff done @newlanternai

exciting news soon tba

https://t.co/tA8gREVTsg

[Prem Seetharaman @pseetharaman]

Well, Jeff Winger talked about something I worked on (Project SuperSonic) - I'm happy. Thanks @joelmchale!

[Adobe Creative Cloud @creativecloud]

Everyone knows the highlight of day two at #AdobeMAX is SNEAKS. 🤫 @joelmchale has the rundown on our exciting new projects currently in development, plus a fairly decent impression of @Draplin 👏 👏 👏

Get more info on these Sneaks: https://t.co/wEG5m0Jd7Q

[Eric Addison @EricAddison]

#ProjectSuperSonic at #AdobeMAX #Sneaks is pretty amazing!

[Justin Salamon @justin_salamon]

What a thrill to present on the big stage! So excited to reveal our Sounds Effects GenAI tech in #ProjectSuperSonic #AdobeMAX

Text-to-SFX and *VOICE*-to-SFX for expressive control!

Huge kudos to @urinieto @pseetharaman @hugggof and our collaborators in design & prototyping!

[Prem Seetharaman @pseetharaman]

My first feature in Premiere Pro, from my time so far with @AdobeResearch, @justin_salamon, @urinieto. Taking interns for Summer 2025!

[Justin Salamon @justin_salamon]

Adobe just announced Generative Extend for Premiere Pro (beta) at #AdobeMAX! Use GenAI to extend your video clip *including the audio*

@pseetharaman @urinieto and me in the Sound Design AI Group at @AdobeResearch worked on the audio part and we're so excited to see it go out!

[Zachary Novack @zacknovack]

Ultra-fast text-to-music generation w/o degrading quality?

Introducing Presto! Distilling Steps and Layers for Accelerating Music Generation

🎹: https://t.co/kTTAYKKtTU
📖: https://t.co/Newhxe6lI6
w/@__gzhu__ @CasebeerJonah @BergKirkpatrick @McAuleyLabUCSD @NicholasJBryan

🧵

[Prem Seetharaman @pseetharaman]

@soumithchintala Thank you!

[Prem Seetharaman @pseetharaman]

Anyone know more about this? @soumithchintala

[Oriol (Uri) Nieto @urinieto]

Wow, torchaudio is no longer maintained?? https://t.co/XLhFnwIeaY I wonder if there's any news on this. This post if 6 months old...