EESEN: End-to-end speech recognition using deep RNN models and WFST-based decoding.

Yajie Miao, Mohammad Gowayyed, Florian Metze

Browse the full ASRU paper archive.