A data-driven caption for L2 listening. Ling. Proc. Ser. 2020;11(1):169-172. doi:10.36505/ExLing-2020/11/0042/000457