Dr. Michael Picheny

Biography

Michael Picheny has worked in the Speech Recognition domain since 1981 after completing his doctorate at MIT. He was heavily involved in the development of IBM’s recognition systems, contributing significantly to the world's first real-time large vocabulary discrete system introduced in 1984 and a range of IBM's product lines in telephony and embedded systems throughout the 1990s. His responsibilities included implementing a suite of Speech Services for both Speech Recognition and Speech Synthesis during his tenure at IBM’s Watson Group. Michael has published numerous papers in journals and conferences regarding various facets of speech recognition. As a Master Inventor at IBM, he co-holds 50 patents and was recognized as such from 1995 to 2000. In addition to his professional commitments, he has volunteered as an Adjunct Professor in the Electrical Engineering Department at Columbia University, where he co-taught a course focused on speech recognition. He is a Fellow of IEEE and ISCA. After a successful 35-year career at IBM, where he led the Speech team in Yorktown Heights, he retired and transitioned to a part-time role as a Research Professor at NYU’s Courant Computer Science Center, where he aims to continue research in speech recognition, particularly addressing challenges associated with accented and disfluent speech, rapid domain adaptation, and cross-modality synergies involving text and vision.

Research Interests