SLP-P27.4
AN EFFECTIVE MIXTURE-OF-EXPERTS APPROACH FOR CODE-SWITCHING SPEECH RECOGNITION LEVERAGING ENCODER DISENTANGLEMENT
Tzu-Ting Yang, Hsin-Wei Wang, Yi-Cheng Wang, National Taiwan Normal University, Taiwan; Chi-Han Lin, E.SUN Financial Holding Co., Ltd., Taiwan; Berlin Chen, National Taiwan Normal University, Taiwan
Session:
SLP-P27: Acoustic modeling for automatic speech recognition Poster
Track:
Speech and Language Processing
Location:
Poster Zone 1B
Poster Board PZ-1B.4
Poster Board PZ-1B.4
Presentation Time:
Thu, 18 Apr, 16:30 - 18:30 (UTC +9)
Session Co-Chairs:
Naoyuki Kanda, Microsoft and Takaaki Hori, Apple
Session SLP-P27
SLP-P27.1: G2PU: Grapheme-to-Phoneme Transducer with Speech Units
Heting Gao, Mark Hasegawa-Johnson, University of Illinois at Urbana-Champaign, United States of America; Chang D. Yoo, Korea Advanced Institute of Science and Technology, Korea, Republic of
SLP-P27.2: IMPROVING KINYARWANDA SPEECH RECOGNITION VIA SEMI-SUPERVISED LEARNING
Antoine Nzeyimana, University of Massachusetts Amherst, United States of America
SLP-P27.3: Improving Oral Reading Fluency Assessment through Sub-sequence Matching of Acoustic Word Embeddings
Yihao Wang, Zhongdi Wu, Southern Methodist University, United States of America; Joseph Nese, University of Oregon, United States of America; Akihito Kamata, Vedant Nilabh, Eric Larson, Southern Methodist University, United States of America
SLP-P27.4: AN EFFECTIVE MIXTURE-OF-EXPERTS APPROACH FOR CODE-SWITCHING SPEECH RECOGNITION LEVERAGING ENCODER DISENTANGLEMENT
Tzu-Ting Yang, Hsin-Wei Wang, Yi-Cheng Wang, National Taiwan Normal University, Taiwan; Chi-Han Lin, E.SUN Financial Holding Co., Ltd., Taiwan; Berlin Chen, National Taiwan Normal University, Taiwan
SLP-P27.5: CHUNKED ATTENTION-BASED ENCODER-DECODER MODEL FOR STREAMING SPEECH RECOGNITION
Mohammad Zeineldeen, Albert Zeyer, Ralf Schlueter, Hermann Ney, RWTH Aachen University / AppTek, Germany
SLP-P27.6: LESS PEAKY AND MORE ACCURATE CTC FORCED ALIGNMENT BY LABEL PRIORS
Ruizhe Huang, Johns Hopkins University, United States of America; Xiaohui Zhang, Zhaoheng Ni, Meta, China; Li Sun, Boston University, United States of America; Moto Hira, Jeff Hwang, Vimal Manohar, Vineel Pratap, Meta, United States of America; Matthew Wiesner, Johns Hopkins University, United States of America; Shinji Watanabe, Carnegie Mellon University, United States of America; Daniel Povey, Xiaomi Corp., China; Sanjeev Khudanpur, Johns Hopkins University, China
SLP-P27.7: EXPLORING ADAPTERS WITH CONFORMERS FOR CHILDREN'S AUTOMATIC SPEECH RECOGNITION
Thomas Rolland, Alberto Abad, INESC-ID, Portugal
SLP-P27.8: IMPROVED CHILDREN'S AUTOMATIC SPEECH RECOGNITION COMBINING ADAPTERS AND SYNTHETIC DATA AUGMENTATION
Thomas Rolland, Alberto Abad, INESC-ID, Portugal
SLP-P27.9: Multi-stream Acoustic Modelling using Raw Real and Imaginary Parts of the Fourier Transform
Erfan Loweimi, University of Cambridge, United Kingdom of Great Britain and Northern Ireland; Zhengjun Yue, Technische Universiteit Delft, Netherlands; Peter Bell, University of Edinburgh, United Kingdom of Great Britain and Northern Ireland; Steve Renals, The University of Edinburgh, United Kingdom of Great Britain and Northern Ireland; Zoran Cvetkovic, King's College London, United Kingdom of Great Britain and Northern Ireland
SLP-P27.10: FLEXIBLE KEYWORD SPOTTING BASED ON HOMOGENEOUS AUDIO-TEXT EMBEDDING
Kumari Nishu, Minsik Cho, Paul Dixon, Devang Naik, Apple, United States of America
Contacts