Technical Program

HLT-L1: Resource-limited Speech Recognition

Session Type: Lecture
Time: Wednesday, March 23, 08:30 - 10:30
Location: Room 3E
Session Chairs: Florian Metze, Carnegie Mellon University and Mari Ostendorf, University of Washington
 
HLT-L1.1: PERSONALIZED SPEECH RECOGNITION ON MOBILE DEVICES
         Ian McGraw; Google Inc.
         Rohit Prabhavalkar; Google Inc.
         Raziel Alvarez; Google Inc.
         Montse Gonzalez Arenas; Google Inc.
         Kanishka Rao; Google Inc.
         David Rybach; Google Inc.
         Ouais Alsharif; Google Inc.
         Hasim Sak; Google Inc.
         Alexander Gruenstein; Google Inc.
         Francoise Beaufays; Google Inc.
         Carolina Parada; Google Inc.
 
HLT-L1.2: LEARNING COMPACT RECURRENT NEURAL NETWORKS
         Zhiyun Lu; University of California, Los Angeles
         Vikas Sindhwani; Google Inc.
         Tara Sainath; Google Inc.
 
HLT-L1.3: TOWARDS IMPLICIT COMPLEXITY CONTROL USING VARIABLE-DEPTH DEEP NEURAL NETWORKS FOR AUTOMATIC SPEECH RECOGNITION
         Shawn Tan; National University of Singapore
         Khe Chai Sim; National University of Singapore
 
HLT-L1.4: ON THE COMPRESSION OF RECURRENT NEURAL NETWORKS WITH AN APPLICATION TO LVCSR ACOUSTIC MODELING FOR EMBEDDED SPEECH RECOGNITION
         Rohit Prabhavalkar; Google Inc.
         Ouais Alsharif; Google Inc.
         Antoine Bruguier; Google Inc.
         Ian McGraw; Google Inc.
 
HLT-L1.5: INVESTIGATING TECHNIQUES FOR LOW RESOURCE CONVERSATIONAL SPEECH RECOGNITION
         Antoine Laurent; Vocapia Research
         Thiago Fraga-Silva; Vocapia Research
         Lori Lamel; LIMSI-CNRS
         Jean-Luc Gauvain; LIMSI-CNRS
 
HLT-L1.6: DATA SELECTION FOR NOISE ROBUST EXEMPLAR MATCHING
         Emre Yilmaz; Katholieke Universiteit Leuven
         Jort F. Gemmeke; Katholieke Universiteit Leuven
         Hugo Van Hamme; Katholieke Universiteit Leuven
 

ICASSP 2016 Patrons