ICASSP 2025 was fun! I organized the Generative Data Augmentation (GenDA) workshop along with my colleagues (Dinesh Manocha at U. of Maryland, Johan Hershey at Google, and Trausti Kristjansson at Amazon). Two students in my group also led the challenges as the task captains (Jaesung Bae and Jackie Lin), which got a lot of attention, too. Nick Bryan (Adobe Research), Apoorv Vyas (Meta), and Fabio Antonacci (Polytechnic University of Milan) gave wonderful keynote talks on music, speech, and audio, respectively.
I also presented a paper that looks back at the 40 years of innovations in perceptual audio coding, co-authored by Jürgen Herre, Schuyler Queckenback, and Jan Skoglund to celebrate the 50th anniversary of ICASSP within the AASP TC. It was great to chair this special session (along with Shoko Araki) as well, where we also hosted 5 other representatives from DCASE, music, source separation, spatial audio, and microphone array processing.
I had the opportunity to discuss the future of AASP TC at the TC meeting and the general technical directions of SPS at the technical directions board meeting. But the most fun part was to interact with global experts in signal processing at the technical sessions, breaks, meals, and bars!