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-48: Speech Recognition 18: Low Resource ASR

Session Type: Poster
Time: Friday, 11 June, 11:30 - 12:15
Location: Gather.Town
Session Chair: Shinji Watanabe, Johns Hopkins University
 
SPE-48.1: MIXSPEECH: DATA AUGMENTATION FOR LOW-RESOURCE AUTOMATIC SPEECH RECOGNITION
         Linghui Meng; Institute of Automation, Chinese Academy of Sciences
         Jin Xu; Institute for Interdisciplinary Information Sciences, Tsinghua University
         Xu Tan; Microsoft Research Asia
         Jindong Wang; Microsoft Research Asia
         Tao Qin; Microsoft Research Asia
         Bo Xu; Institute of Automation, Chinese Academy of Sciences
 
SPE-48.2: END-TO-END MULTILINGUAL AUTOMATIC SPEECH RECOGNITION FOR LESS-RESOURCED LANGUAGES: THE CASE OF FOUR ETHIOPIAN LANGUAGES
         Solomon Teferra Abate; Addis Ababa University
         Martha Yifiru Tachbelie; Addis Ababa University
         Tanja Schultz; University of Bremen
 
SPE-48.3: IMPROVED DATA SELECTION FOR DOMAIN ADAPTATION IN ASR
         Shannon Wotherspoon; Raytheon BBN
         William Hartmann; Raytheon BBN
         Matthew Snover; Raytheon BBN
         Owen Kimball; Raytheon BBN
 
SPE-48.4: BI-APC: BIDIRECTIONAL AUTOREGRESSIVE PREDICTIVE CODING FOR UNSUPERVISED PRE-TRAINING AND ITS APPLICATION TO CHILDREN’S ASR
         Ruchao Fan; University of California, Los Angeles
         Amber Afshan; University of California, Los Angeles
         Abeer Alwan; University of California, Los Angeles
 
SPE-48.5: META-ADAPTER: EFFICIENT CROSS-LINGUAL ADAPTATION WITH META-LEARNING
         Wenxin Hou; Tokyo Institute of Technology
         Yidong Wang; Tokyo Institute of Technology
         Shengzhou Gao; Tokyo Institute of Technology
         Takahiro Shinozaki; Tokyo Institute of Technology
 
SPE-48.6: ERROR-DRIVEN FIXED-BUDGET ASR PERSONALIZATION FOR ACCENTED SPEAKERS
         Abhijeet Awasthi; IIT Bombay
         Aman Kansal; IIT Bombay
         Sunita Sarawagi; IIT Bombay
         Preethi Jyothi; IIT Bombay