He does the full audio range; I do only the stuff below about 10KHz and text - Natural Language Processing
Ben's currently doing Topic Modelling at Metabob - and managing interns and university collaborations, which means I'm learning from them. Recent work includes the detection of clinical depression in speech at Ellipsis Health in San Francisco, and synthesis of a specific voice using augmented training of deep neural networks using a limited number of target speaker recordings. His experience with interactive speech applications spans 30 years in Japan and California at Toyota ITC , Nuance Communications ,
Ben's daughter, Grace , graduated Waseda University and works at Disney Animation in beautiful downtown Burbank.