15 citations · 15 across the 4 of their papers we have counts for
1 paper · 1 filter
Xiuwen Zheng, Bornali Phukon, Mark Hasegawa-Johnson
This paper enhances dysarthric and dysphonic speech recognition by fine-tuning pretrained automatic speech recognition (ASR) models on the 2023-10-05 data package of the Speech Acc…