Skip to main content
. Author manuscript; available in PMC: 2019 Jul 1.
Published in final edited form as: Proc Conf Assoc Comput Linguist Meet. 2018 Jul;2018:2225–2235.

Figure 2:

Figure 2:

Fusion strategies. t_hi: word-level textual bidirectional state. t_αi: word-level textual attention distribution. w_hi: word-level acoustic bidirectional state. w_αi: word-level acoustic attention distribution. s_αi: shared attention distribution. u_αi: fine-tuning attention distribution. Vi: shared word-level representation.