Sequence-to-Sequence Asr Optimization Via Reinforcement Learning.

Andros Tjandra, Sakriani Sakti, Satoshi Nakamura

Browse the full ICASSP paper archive.