构象异构
编码器
变压器
计算机科学
机器翻译
语言模型
人工智能
自然语言处理
语音识别
工程类
物理
电气工程
量子力学
电压
分子
操作系统
作者
J.Q. Ning,Yugang Dai,Guanyu Li,Senyan Li,Sirui Li,Guangming Li
摘要
Deep learning has led to remarkable success in many research fields,such as machine translation,speech recognition. Combined with the characteristics of the Tibetan and the current development of Tibetan, with the help of transformer, conformer and other frameworks, we studies how changes in modeling units and encoder types affect Tibetan speech recognition. The modeling unit in this paper generates subwords through bpe algorithm or automatically divides Tibetan characters, and the encoder types are transformer and conformer. Finally, we verify it on the open-source Tibetan language dataset xbmu-amdo31. The experimental results show that the Tibetan characters as the modeling unit and conformer as the encoder has the best performance. The CER of conformer encoder whose modeling unit is character is 11.89%, which is 44.51% lower than the transformer model. The WER of conformer encoder whose modeling unit is subword is 17.40%, which is relatively reduced by 32.27%.
科研通智能强力驱动
Strongly Powered by AbleSci AI