Technical Program

SLP-L2: End-to-end Speech Recognition II: New Models

Session Type: Lecture
Time: Tuesday, May 14, 17:30 - 19:30
Location: Auditorium 1
Session Chairs: Jinyu Li, Microsoft and Yanzhang He, Google Inc.
 
SLP-L2.1: A SPELLING CORRECTION MODEL FOR END-TO-END SPEECH RECOGNITION
Manuscript Link:  Click here to view manuscript on IEEE Xplore
         Jinxi Guo; University of California, Los Angeles
         Tara N. Sainath; Google, Inc.
         Ron J. Weiss; Google, Inc.
 
SLP-L2.2: SELF-ATTENTION ALIGNER: A LATENCY-CONTROL END-TO-END MODEL FOR ASR USING SELF-ATTENTION NETWORK AND CHUNK-HOPPING
Manuscript Link:  Click here to view manuscript on IEEE Xplore
         Linhao Dong; Institute of Automation, Chinese Academy of Sciences
         Feng Wang; Institute of Automation, Chinese Academy of Sciences
         Bo Xu; Institute of Automation, Chinese Academy of Sciences
 
SLP-L2.3: LARGE CONTEXT END-TO-END AUTOMATIC SPEECH RECOGNITION VIA EXTENSION OF HIERARCHICAL RECURRENT ENCODER-DECODER MODELS
Manuscript Link:  Click here to view manuscript on IEEE Xplore
         Ryo Masumura; NTT Corporation
         Tomohiro Tanaka; NTT Corporation
         Takafumi Moriya; NTT Corporation
         Yusuke Shinohara; NTT Corporation
         Takanobu Oba; NTT Corporation
         Yushi Aono; NTT Corporation
 
SLP-L2.4: TRIGGERED ATTENTION FOR END-TO-END SPEECH RECOGNITION
Manuscript Link:  Click here to view manuscript on IEEE Xplore
         Niko Moritz; Mitsubishi Electric Research Laboratories
         Takaaki Hori; Mitsubishi Electric Research Laboratories
         Jonathan Le Roux; Mitsubishi Electric Research Laboratories
 
SLP-L2.5: ON USING 2D SEQUENCE-TO-SEQUENCE MODELS FOR SPEECH RECOGNITION
Manuscript Link:  Click here to view manuscript on IEEE Xplore
         Parnia Bahar; RWTH Aachen University
         Albert Zeyer; RWTH Aachen University
         Ralf Schlüter; RWTH Aachen University
         Hermann Ney; RWTH Aachen University
 
SLP-L2.6: CRF-BASED SINGLE-STAGE ACOUSTIC MODELING WITH CTC TOPOLOGY
Manuscript Link:  Click here to view manuscript on IEEE Xplore
         Hongyu Xiang; Tsinghua University
         Zhijian Ou; Tsinghua University