Jaekwon Im

PhD Student @ Music and Audio Computing Lab, KAIST

[Keunwoo Choi @keunwoochoi]

https://t.co/YP8kYnwCiR
카이스트에서 진행한 유하영 재즈워크샵 영상이 올라왔습니다. 아주 좋은 질문으로 가득한 좋은 시간이었네요.

진행 @osalooloo
주최 @juhan_nam
기획 @keunwoochoi

[Jaekwon Im @osalooloo]

I'm attending #ICASSP2025 and presenting FlashSR. I’d love to connect with anyone interested in audio super-resolution, audio generation, diffusion, flow matching, or related topics. Please feel free to stop by the session or message me for a coffee chat.

[Jaekwon Im @osalooloo]

🌟 Excited to announce the release of the code and model weights for FlashSR: One-step Versatile Audio Super-resolution via Diffusion Distillation, accepted at ICASSP 2025! 🎉

🔗 Check out the demo, code, and paper here: https://t.co/I2JfTmWsHl

[Jaekwon Im @osalooloo]

⚡ Achieves performance approximately 14 times faster than real-time on a single A6000 GPU.
🔬 Applies diffusion distillation to the audio super-resolution task, and introduces the SR Vocoder, specifically designed for SR models operating on mel-spectrograms.

[Jaekwon Im @osalooloo]

🌟 Excited to announce the release of the code and model weights for FlashSR: One-step Versatile Audio Super-resolution via Diffusion Distillation, accepted at ICASSP 2025! 🎉

🔗 Check out the demo, code, and paper here: https://t.co/I2JfTmWsHl

[Jaekwon Im @osalooloo]

🚀 A one-step diffusion model for audio super-resolution that can upsample various types of audio—music, speech, and sound effects—from any sampling rate between 4kHz and 32kHz to 48kHz.

[Jaekwon Im @osalooloo]

Key highlights of FlashSR:

[Keunwoo Choi @keunwoochoi]

hi music people, i wrote a tutorial on large language models and music information retrieval. of course it's called..

LLMs <3 MIR

🥁 have fun!

https://t.co/g7VXyW0lgd

[이준원 Junwon Lee @jnwnlee]

Are you wondering how to make a Synchronized Foley Sound for Sora-made videos? Here’s our work 🔊T-Foley. (ICASSP 2024)

Demo https://t.co/0dm6WUXymQ
Code https://t.co/48OyVRFzaZ

(It was a pleasure to work with @yo__j__)

[Jaekwon Im @osalooloo]

Super excited to share my work "DIFFRENT: A Diffusion Model for Recording Environment Transfer of Speech" at #ICASSP2024 on April 17th.
Looking forward to connecting with many researchers :).
I'm always up for a coffee chat too!

Demo: https://t.co/PjgAQMHWzT

[Christian Steinmetz @csteinmetz1]

DIFFRENT: A Diffusion Model for Recording Environment Transfer of Speech

Proposes a diffusion model to transfer the environment from a reference sound to a new recording taking into account the microphone and placement, room acoustics, and ambient noise.

https://t.co/nZ5vALZdOH

[Jaekwon Im @osalooloo]

@csteinmetz1 Thank you so much for sharing my work!!😀😀

[Keunwoo Choi @keunwoochoi]

https://t.co/z533CBcKuR
Our paper about DCASE Challenge T7 - Foley Sound Synthesis was accepted to the DCASE Workshop 🥳

I can't make it to Finland🇫🇮, but some of the authors will be there to tell you what we went through while organizing the first generative challenge at DCASE.

[AuditoryLab @AuditoryLabInfo]

Worked hard, learned a lot, and met a great team during the first-ever competition for generative everyday sounds at @DCASE_Challenge, Task 7 Foley sound! @keunwoochoi @osalooloo @KeisukeImoto @forthshinji @YukiOkamoto19 Mathieu Lagrange, Brian McFee.

[arXiv Sound @ArxivSound]

``Foley Sound Synthesis at the DCASE 2023 Challenge. (arXiv:2304.12521v1 [https://t.co/mPAjntoGrG]),'' Keunwoo Choi, Jaekwon Im, Laurie Heller, Brian McFee, Keisuke Imoto, Yuki Okamoto, Mathieu Lagrange, Shinosuke Takamichi,
https://t.co/pOr7wZp1R8

[AuditoryLab @AuditoryLabInfo]

Organizers:
Keunwoo Choi, Jaekwon Im, Gaudio Labs
Keisuke Imoto, Doshisha U
Mathieu Lagrange, CNRS/Ecole Central de Nantes
Laurie Heller, CMU
Brian McFee, NYU
Yuki Okamoto, Ritsumeikan U
Shinnosuke Takamichi, The U of Tokyo
@KeisukeImoto @keunwoochoi @forthshinji @osalooloo

[AuditoryLab @AuditoryLabInfo]

Just launched a generative audio challenge for environmental sounds! Synthesis by ML or any method. Deadline May 15.
Task 7 of @DCASE_Challenge 2023. https://t.co/rk9axDvztW
@ieee_AASP @acousticsorg @AESorg @kaggle #foleysynthesischallenge #audiosynthesis #soundclassification