Integrating Knowledge into End-to-End Speech Recognition from External Text-Only Data

Abstract

Attention-based encoder-decoder (AED) models have achieved promising performance in speech recognition. However, because of the end-to-end training, an AED model is usually trained with speech-text paired data. It is challenging to incorporate external text-only data into AED models. Another issue of the AED model is that it does not use the right context of a text token while predicting the token. To alleviate the above two issues, we propose a unified method called LST (Learn Spelling from Teachers) to integrate knowledge into an AED model from the external text-only data and leverage the whole context in a sentence. The method is divided into two stages. First, in the representation stage, a language model is trained on the text. It can be seen as that the knowledge in the text is compressed into the LM. Then, at the transferring stage, the knowledge is transferred to the AED model via teacher-student learning. To further use the whole context of the text sentence, we propose an LM called causal cloze completer (COR), which estimates the probability of a token, given both the left context and the right context of it. Therefore, with LST training, the AED model can leverage the whole context in the sentence. Different from fusion based methods, which use LM during decoding, the proposed method does not increase any extra complexity at the inference stage. We conduct experiments on two scales of public Chinese datasets AISHELL-1 and AISHELL-2. The experimental results demonstrate the effectiveness of leveraging external text-only data and the whole context in a sentence with our proposed method, compared with baseline hybrid systems and AED model based systems.

arXiv commentSubmitted TASLP

References in corpus (15)

• Distilling the Knowledge in a Neural Network
• SpecAugment: A Simple Data Augmentation Method for Automatic Speech Recognition
• XLNet: Generalized Autoregressive Pretraining for Language Understanding
• Attention-Based Models for Speech Recognition
• Scheduled Sampling for Sequence Prediction with Recurrent Neural Networks
• Sequence Transduction with Recurrent Neural Networks
• A Comparative Study on Transformer vs RNN in Speech Applications
• On Using Monolingual Corpora in Neural Machine Translation
• Regularizing Neural Networks by Penalizing Confident Output Distributions
• AISHELL-2: Transforming Mandarin ASR Research Into Industrial Scale
• Language Modeling with Deep Transformers
• A Comprehensive Study of Deep Bidirectional LSTM RNNs for Acoustic Modeling in Speech Recognition
• Recent Developments on ESPnet Toolkit Boosted by Conformer
• Unsupervised pre-training for sequence to sequence speech recognition
• CAT: CRF-based ASR Toolkit