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-13: Speech Recognition 5: New Algorithms

Session Type: Poster
Time: Wednesday, 9 June, 13:00 - 13:45
Location: Gather.Town
 
SPE-13.1: META-LEARNING FOR IMPROVING RARE WORD RECOGNITION IN END-TO-END ASR
         Florian Lux; University of Stuttgart
         Ngoc Thang Vu; University of Stuttgart
 
SPE-13.2: A COMPARISON OF METHODS FOR OOV-WORD RECOGNITION ON A NEW PUBLIC DATASET
         Rudolf A Braun; Idiap Research Institute
         Srikanth Madikeri; Idiap Research Institute
         Petr Motlicek; Idiap Research Institute
 
SPE-13.3: CONVOLUTIONAL DROPOUT AND WORDPIECE AUGMENTATION FOR END-TO-END SPEECH RECOGNITION
         Hainan Xu; Google
         Yinghui Huang; Google
         Yun Zhu; Google
         Kartik Audhkhasi; Google
         Bhuvana Ramabhadran; Google
 
SPE-13.4: PARTIALLY OVERLAPPED INFERENCE FOR LONG-FORM SPEECH RECOGNITION
         Tae Gyoon Kang; Samsung Electronics
         Ho-Gyeong Kim; Samsung Electronics
         Min-Joong Lee; Samsung Electronics
         Jihyun Lee; Samsung Electronics
         Hoshik Lee; Samsung Electronics
 
SPE-13.5: FOCUS ON THE PRESENT: A REGULARIZATION METHOD FOR THE ASR SOURCE-TARGET ATTENTION LAYER
         Nanxin Chen; Johns Hopkins University
         Piotr Zelasko; Johns Hopkins University
         Jesús Villalba; Johns Hopkins University
         Najim Dehak; Johns Hopkins University
 
SPE-13.6: BIFOCAL NEURAL ASR: EXPLOITING KEYWORD SPOTTING FOR INFERENCE OPTIMIZATION
         Jon Macoskey; Amazon.com
         Grant Strimel; Amazon.com
         Ariya Rastrow; Amazon.com