Technical Program
SP-L5: Acoustic Modeling for Speech Recognition |
Session Type: Lecture |
Time: Wednesday, March 23, 16:00 - 18:00 |
Location: Room 3E |
Session Chairs: Geoffrey Zweig, Microsoft Research and Shinji Watanabe, Mitsubishi Electric Research Labs |
SP-L5.1: ON TRAINING THE RECURRENT NEURAL NETWORK ENCODER-DECODER FOR LARGE VOCABULARY END-TO-END SPEECH RECOGNITION |
Liang Lu; The University of Edinburgh |
Xingxing Zhang; The University of Edinburgh |
Steve Renals; The University of Edinburgh |
SP-L5.2: DISCRIMINATIVELY TRAINED JOINT SPEAKER AND ENVIRONMENT REPRESENTATIONS FOR ADAPTATION OF DEEP NEURAL NETWORK ACOUSTIC MODELS |
Maofan Yin; Shanghai Jiao Tong University |
Sunil Sivadas; Institute for Infocomm Research |
Kai Yu; Shanghai Jiao Tong University |
Bin Ma; Institute for Infocomm Research |
SP-L5.3: A COMPARISON BETWEEN DEEP NEURAL NETS AND KERNEL ACOUSTIC MODELS FOR SPEECH RECOGNITION |
Zhiyun Lu; University of California, Los Angeles |
Dong Guo; University of Southern California |
Alireza Bagheri Garakani; University of Southern California |
Kuan Liu; University of Southern California |
Avner May; Columbia University |
Aurelien Bellet; Team Magnet, INRIA Lille - Nord Europe |
Linxi Fan; Columbia University |
Michael Collins; Columbia University |
Brian Kingsbury; IBM |
Michael Picheny; IBM |
Fei Sha; University of California, Los Angeles |
SP-L5.4: FACTORED SPATIAL AND SPECTRAL MULTICHANNEL RAW WAVEFORM CLDNNS |
Tara Sainath; Google Inc. |
Ron Weiss; Google Inc. |
Kevin Wilson; Google Inc. |
Arun Narayanan; Google Inc. |
Michiel Bacchiani; Google Inc. |
SP-L5.5: HOW NEURAL NETWORK FEATURES AND DEPTH MODIFY STATISTICAL PROPERTIES OF HMM ACOUSTIC MODELS |
Suman Ravuri; International Computer Science Institute; University of California - Berkeley |
Steven Wegmann; International Computer Science Institute/Semantic Machines |
SP-L5.6: LINEARLY AUGMENTED DEEP NEURAL NETWORK |
Pegah Ghahremani; Johns Hopkins University |
Jasha Droppo; Microsoft Research |
Michael L. Seltzer; Microsoft Research |