Inventor
CHANG SHUO-YIIN
US14 patents
Patents
14 patentsUS11475880B2Oct 18, 2022
Joint endpointing and automatic speech recognition
GOOGLE LLC2 citations71
US10929754B2Feb 23, 2021
Unified endpointer using multitask and multidomain learning
GOOGLE LLC4 citations71
US11676625B2Jun 13, 2023
Unified endpointer using multitask and multidomain learning
GOOGLE LLC0 citations61
US12548561B2Feb 10, 2026
Universal monolingual output layer for multilingual speech recognition
GOOGLE LLC0 citations51
US12354597B2Jul 8, 2025
Disfluency detection models for natural conversational voice systems
GOOGLE LLC0 citations51
US12094453B2Sep 17, 2024
Fast emit low-latency streaming ASR with sequence-level emission regularization utilizing forward and backward probabilities between nodes of an alignment lattice
GOOGLE LLC0 citations51
US12073824B2Aug 27, 2024
Two-pass end to end speech recognition
GOOGLE LLC0 citations51
US12211509B2Jan 28, 2025
Fusion of acoustic and text representations in RNN-T
GOOGLE LLC0 citations50
US12525233B2Jan 13, 2026
Joint segmenting and automatic speech recognition
GOOGLE LLC0 citations48
US12525228B2Jan 13, 2026
Unified end-to-end speech recognition and endpointing using a switch connection
GOOGLE LLC0 citations48
US12183322B2Dec 31, 2024
Language agnostic multilingual end-to-end streaming on-device ASR system
GOOGLE LLC0 citations48
US12586579B2Mar 24, 2026
End-to-end segmentation in a two-pass cascaded encoder automatic speech recognition model
GOOGLE LLC0 citations47
US12315497B2May 27, 2025
Intended query detection using E2E modeling for continued conversation
GOOGLE LLC0 citations47
US12322383B2Jun 3, 2025
Predicting word boundaries for on-device batching of end-to-end speech recognition models
GOOGLE LLC0 citations46