Jason Pelecanos  Jason Pelecanos photo       

contact information

Speaker Recognition and Audio Analytics Research
Thomas J. Watson Research Center, Yorktown Heights, NY USA



Dr Jason Pelecanos earned his PhD at the Queensland University of Technology in 2003 in the area of signals processing. The thesis was titled “Robust Automatic Speaker Recognition”. From 2003-2006 he was a post-doctoral researcher at the IBM T.J. Watson Research Center. He is currently a Research Staff Member at IBM Research with a core focus on automatic speaker and language recognition technology and audio analytics. In 2009, as part of a team, he received an IBM Technical Achievement Award for advancing the field of “Conversational Biometrics” and an IEEE best system paper award in 2007 for “eChronicling”. He has also worked on and/or headed several government research contracts on automatic speaker recognition. He is known for mismatch compensation techniques, namely feature warping and Gaussianization, which address channel and additive noise effects. He has published several works on topics ranging from conversational biometrics to feature compensation and emerging non-linear modeling techniques. He has participated in NIST speaker recognition evaluations, has served on the NSF grants review panel for speech, is a member of the IEEE SLTC and is a reviewer for multiple IEEE journals and conferences.