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 |