Acoustic modeling with deep neural networks using raw time signal for LVCSR.

Zoltn Tske, Pavel Golik, Ralf Schlter, Hermann Ney

Browse the full Interspeech paper archive.