Program
09:30 – 10:00
Registration Deck Open (Refreshments Available)
10:00 - 10:10
Opening Remarks
10:10 - 10:55
Keynote
A journey into self-supervised learning for speech processing
Michael Auli
10:55 - 11:55
Invited Talks
10:55 – 11:10
Improving Self-supervised Speech Pre-training via Model-level Consistency Regularization
Nam Soo Kim
11:10 – 11:25
Self-supervised predictive coding models encode speaker and phonetic information in orthogonal subspaces
Hao Tang
11:25 – 11:40
MASR: Multi-label Aware Speech Representation
Shikhar Vashishth
11:40 – 11:55
Combined Q&A Session for the three speakers
11:55 – 12:15
11:55 – 12:10
Introduction to AV-SUPERB
Shang-Wen (Daniel) Li, Yuan Tseng
12:10 - 12:15
Q&A Session
12:15 - 13:15
Lunch Break (lunch provided by workshop)
13:15 – 14:15
Poster Session
14:15 - 15:00
14:15 – 14:25
Introduction to ML-SUPERB and the ML-SUPERB Challenge
Shinji Watanabe, Jiatong Shi
14:25 – 14:35
Virtual Presentation of ML-SUPERB Challenge Winner
Hongfei Xue
14:35 – 14:45
ML-SUPERB Data Track Highlights on Taiwanese and Quechua
Kalvin Chang, William Chen
14:45 – 15:00
Combined Q&A Segment
15:00 – 15:45
15:00 – 15:15
Introduction to SLUE
Karen Livescu
15:15 – 15:30
SLUE-PERB: A Spoken Language Understanding Performance Benchmark and Toolkit
Roshan Sharma
15:30 – 15:45
Combined Q&A Segment
15:45 – 16:05
Coffee Break
16:05 – 16:50
16:05 – 16:15
Introduction to Dynamic-SUPERB
Hung-yi Lee & Chien-yu Huang
16:15 – 16:30
16:30 – 16:45
16:45 – 16:55
Combined Q&A Segment
16:55 – 17:30
Expert Panel Discussion
Michael Auli, Karen Livescu, Pedro J. Moreno, Bhuvana Ramabhadran, Shinji Watanabe
The attendees can start departing at 17:30 PM to attend the welcome reception