An uncertainty decoding approach to noise- and reverberation-robust speech recognition.

Roland Maas, Akshaya Thippur, Armin Sehr, Walter Kellermann

Browse the full ICASSP paper archive.