AdaST: Dynamically Adapting Encoder States in the Decoder for End-to-End Speech-to-Text Translation.

Wuwei Huang, Dexin Wang, Deyi Xiong

Browse the full ACL paper archive.