T-GSA: Transformer with Gaussian-Weighted Self-Attention for Speech Enhancement.

Jaeyoung Kim Mostafa El-Khamy Jungwon Lee

Published in: ICASSP (2020)

Keyphrases

speech enhancement
noisy environments
signal to noise ratio
noise reduction
single channel
linear prediction
smoothing algorithm
neural network
visual attention
speech signal
pattern recognition
maximum likelihood
wiener filter