Roberto Sicconi, Program Director at IBM T.J. Watson Research, has received his
M.S degree from the Politecnico University of Milan, Italy and his Ph.D. degree
from the same University in 1985 in the area of image processing and new
Computer Aided Design techniques.
He joined IBM Italy in 1985 to work on development of multimedia platforms,
both hardware and software, working on hardware acceleration for hi-resolution
PC and RISC/6000 workstations, DSP-based front-ends for IBM speech recognitions
systems on PCs, PS/2 and laptops. In 1990 he became manager of the Multimedia
Development Lab at Vimercate, Italy, leading development of speech recognition
products (Martin/Voicetype), high-throughput bank check image scanners, a
high-resolution negative film scanner, TV-Teletext VBI receivers, high-speed
modems, a satellite Internet data broadcasting system, IBM cryptographic
systems for data security, a portable videoconferencing system for laptops. In
1998 he moved to the US to work as executive assistant to the GM of IBM
Microelectronics, leading two cooperation projects with IBM Research (a speech
chip and a Bluetooth chip).
Dr. Ing Sicconi has been with the Human Language Technology group at IBM T. J.
Watson Research Center since 2000. He currently manages development of
exploratory prototypes of conversational and multimodal user interfaces in
smartphones and other consumer devices (STB, videogames), with a special focus
on cars, both in standalone and in network-connected configurations.
Current research projects involve advanced conversational dialog systems,
leveraging natural speech input, dynamic dialog management, workload
management, context- and environment-aware communication with the user to
provide ease-to-use and safe access to services, both local and across the
network. Latest additions are visual speech activity detection, combined with
head-pose detection, eye-gaze monitoring and lip-motion feature extraction to
deliver integrated audio-visual speech recognition for improved robustness in
challenging noisy environments.