【论文笔记】Improving Transformer-based End-to-End Speech Recognition with CTC and LM Integration

题目

Improving Transformer-based End-to-End Speech Recognition with Connectionist Temporal Classification and Language Model Integration

链接

http://www.isca-speech.org/archive/Interspeech_2019/abstracts/1938.html

标签

Speech Recognition, Transformer, CTC, LM

Contributions

• 将CTC，LM与Transformer在decoding阶段融合，实现更好的识别表现
• 通过实验调查了融合效果在大数据集上的表现
• 在开源项目ESPnet上实现基于Transformer的ASR toolkit

亮点与启发

文章指出，Transformer应用于ASR主要有两个问题：

1. 相比于 RNN-based ASR，收敛速度较慢。
2. 不易与语言模型结合

而通过与CTC在解码阶段的结合，可以加速Transformer的收敛速度。

CTC can encourage monotonic alignment between the speech and transcription. There