3 papers
cs.LG2024
Federated Learning of Large ASR Models in the Real World
Yonghui Xiao, Yuxin Ding, Changwan Ryu +2
Federated learning (FL) has shown promising results on training machine learning models with privacy preservation. However, for large models with over 100 million parameters, the t…
cs.SD2023
Edit Distance based RL for RNNT decoding
Dongseong Hwang, Changwan Ryu, Khe Chai Sim
RNN-T is currently considered the industry standard in ASR due to its exceptional WERs in various benchmark tests and its ability to support seamless streaming and longform transcr…
cs.LG2021
Enabling On-Device Training of Speech Recognition Models with Federated Dropout
Dhruv Guliani, Lillian Zhou, Changwan Ryu +5
Federated learning can be used to train machine learning models on the edge on local data that never leave devices, providing privacy by default. This presents a challenge pertaini…