2021 IEEE International Conference on Acoustics, Speech and Signal Processing

6-11 June 2021 • Toronto, Ontario, Canada

Extracting Knowledge from Information

2021 IEEE International Conference on Acoustics, Speech and Signal Processing

6-11 June 2021 • Toronto, Ontario, Canada

Extracting Knowledge from Information

Technical Program

SPE-1: Speech Recognition 1: Neural Transducer Models 1

Session Type: Poster
Time: Tuesday, 8 June, 13:00 - 13:45
Location: Gather.Town
Session Chair: Tara Sainath, Google Inc.
 
SPE-1.1: IMPROVING RNN TRANSDUCER MODELING FOR SMALL-FOOTPRINT KEYWORD SPOTTING
         Yao Tian; Bytedance
         Haitao Yao; Bytedance
         Meng Cai; Bytedance
         Yaming Liu; Bytedance
         Zejun Ma; Bytedance
 
SPE-1.2: CASCADED ENCODERS FOR UNIFYING STREAMING AND NON-STREAMING ASR
         Arun Narayanan; Google Inc.
         Tara N. Sainath; Google Inc.
         Ruoming Pang; Google Inc.
         Jiahui Yu; Google Inc.
         Chung-Cheng Chiu; Google Inc.
         Rohit Prabhavalkar; Google Inc.
         Ehsan Variani; Google Inc.
         Trevor Strohman; Google Inc.
 
SPE-1.3: A BETTER AND FASTER END-TO-END MODEL FOR STREAMING ASR
         Bo Li; Google
         Anmol Gulati; Google
         Jiahui Yu; Google
         Tara N. Sainath; Google
         Chung-Cheng Chiu; Google
         Arun Narayanan; Google
         Shuo-Yiin Chang; Google
         Ruoming Pang; Google
         Yanzhang He; Google
         James Qin; Google
         Wei Han; Google
         Qiao Liang; Google
         Yu Zhang; Google
         Trevor Strohman; Google
         Yonghui Wu; Google
 
SPE-1.4: EFFICIENT KNOWLEDGE DISTILLATION FOR RNN-TRANSDUCER MODELS
         Sankaran Panchapagesan; Google, LLC
         Daniel Park; Google, LLC
         Chung-Cheng Chiu‎; Google, LLC
         Yuan Shangguan; Facebook, Inc.
         Qiao Liang; Google, LLC
         Alexander Gruenstein; Google, LLC
 
SPE-1.5: PHONEME BASED NEURAL TRANSDUCER FOR LARGE VOCABULARY SPEECH RECOGNITION
         Wei Zhou; RWTH Aachen University
         Simon Berger; RWTH Aachen University
         Ralf Schlüter; RWTH Aachen University
         Hermann Ney; RWTH Aachen University
 
SPE-1.6: RNN-T BASED OPEN-VOCABULARY KEYWORD SPOTTING IN MANDARIN WITH MULTI-LEVEL DETECTION
         Zuozhen Liu; Key Laboratory of Speech Acoustics and Content Understanding, Institute of Acoustics
         Ta Li; Key Laboratory of Speech Acoustics and Content Understanding, Institute of Acoustics
         Pengyuan Zhang; Key Laboratory of Speech Acoustics and Content Understanding, Institute of Acoustics