WASPAA 2005
Kelly, Dan and I just had our paper at the IEEE biennial Workshop on Applications of Signal Processing to Audio and Acoustics, Learning Auditory Models of Machine Voices. We got the much vaunted “crackpot slot” (last paper on the last day,) an honor I haven’t had since… that time I presented text mining at ICMC I guess. Kelly gave the talk to rapturous applause. Sadly this year WASPAA toned down the music to make room for more variety; usually a good thing but the music-IR regulars weren’t there and I’m not DSP-savvy enough to grok the filters n’ beamforming. As usual it’s always a gorgeous place to go (the Catskills in upstate NY) and catch up with everyone. Tristan also gave a talk on his multi-layer structural similarity stuff and a poster on downbeat prediction.
Julius Smith did a great keynote on the last day with a sweet tribute to Bob Moog — Keith Emerson’s solo in ELP’s “Lucky Man” blaring out of the conference speakers while a nice slideshow of Bob projected onto the big screen. I remembered finding Bob sitting alone at lunch at WASPAA 2001 after his keynote. Everyone assumed he was sitting at a reserved “important person” table (conferences usually have these, intentional or not.) I got the courage to sit there anyway and I barely remember what I said because I was so nervous I’d make a fool of myself regarding some buffet tomatoes on shirt etc etc — during the synthprog moment of remembrance it struck me how we’ve linked a set of timbres to a person instead of a machine. I wish there were more examples of this in synthesis history! I will assume Shifty will be the next one.