Listen and Fill in the Missing Letters: Non-Autoregressive Transformer for Speech Recognition

Understand

Recently very deep transformers have outperformed conventional bi-directional long short-term memory networks by a large margin in speech recognition.

• However, to put it into production usage, inference computation cost is still a serious concern in real scenarios.
• In this paper, we study two different non-autoregressive transformer structure for automatic speech recognition (ASR): A-CMLM and A-FMLM.
• During training, for both frameworks, input tokens fed to the decoder are randomly replaced by special mask tokens.