Time-varying residual noise feature model estimation for multi-microphone speech recognition.

Takuya Yoshioka, Emmanuel Ternon, Tomohiro Nakatani

Browse the full ICASSP paper archive.