Sign in
End-to-End Sound Source Enhancement Using Deep Neural Network in the Modified Discrete Cosine Transform Domain.
Yuma Koizumi
Noboru Harada
Yoichi Haneda
Yusuke Hioka
Kazunori Kobayashi
Published in:
ICASSP (2018)
Keyphrases
</>
end to end
transform domain
discrete cosine
sound source
spatial domain
dct domain
discrete cosine transform
transform coefficients
dct coefficients
pattern recognition
bit plane
distributed video coding
multiscale
audio visual
feature extraction
transform coding
frequency domain
quality assessment
image processing