New IEEE/ACM Journal Publication

28 November 2017, by David Mosteller

Key Contribution: If modern machine learning techniques only model speech spectral envelopes, for speech enhancement super-Gaussian models make an important difference.

Robert Rehr, Timo Gerkmann, "On the Importance of Super-Gaussian Speech Priors for Machine-Learning Based Speech Enhancement", IEEE/ACM Trans. Audio, Speech, Language Proc.