Enhanced Synthesis of Passively Heard Speech from Electrocorticography Signals Using Image-to-Image Spectrogram Translation

Hongsang Lee1, Jihun Hwang2, Kyungjun Kim3, Gyuwon Lee4, Chun Kee Chung5 and Chang-Hwan Im1,2,3*

1Department of Artificial Intelligence, Hanyang University, Republic of Korea

2Department of Electronic Engineering, Hanyang University, Republic of Korea

3Department of Biomedical Engineering, Hanyang University, Republic of Korea

4Department of Brain and Cognitive Sciences, Seoul National University College of Natural Sciences, Republic of Korea

5Neuroscience Research Institute, Seoul National University Medical Research Center, Republic of Korea

P1
Original Initial Synthesis Enhanced
P1
Original Initial Synthesis Enhanced
P2
Original Initial Synthesis Enhanced
P2
Original Initial Synthesis Enhanced