上一条: 日本、韩国、哈萨克、越南普通话学习者的轻声产出
下一条: Towards the Use of Pretrained Language Model GPT-2 for Testing the Hypothesis of Communicative Efficiency in the Lexicon