Learning ASR-Robust Contextualized Embeddings for Spoken Language Understanding
Authors:
Preprint
Abstract
Employing pre-trained language models (LM) to extract contextualized word representations has achieved state-of-the-art performance on various NLP tasks. However, applying this technique to noisy transcripts generated by automatic speech recognizer (ASR) is concerned. Therefore, this paper focuses on making contextualized representations more ASR-robust. We propose a novel confusion-aware fine-tuning method to mitigate the impact of ASR errors to pre-trained LMs. Specifically, we fine-tune LMs to produce similar representations for acoustically confusable words that are obtained from word confusion networks (WCNs) produced by ASR. Experiments on the benchmark ATIS dataset show that the proposed method significantly improves the performance of spoken language understanding when performing on ASR transcripts.Keywords
All Related Versions
This publication has 9 references indexed in Scilit:
- Unsupervised Transfer Learning for Spoken Language Understanding in Intelligent AgentsProceedings of the AAAI Conference on Artificial Intelligence, 2019
- BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding2019
- An Embarrassingly Simple Approach for Transfer Learning from Pretrained Language Models2019
- Slot-Gated Modeling for Joint Slot Filling and Intent Prediction2018
- LatticeRnn: Recurrent Neural Networks Over Lattices2016
- Using Recurrent Neural Networks for Slot Filling in Spoken Language UnderstandingIEEE/ACM Transactions on Audio, Speech, and Language Processing, 2014
- Semantic parsing using word confusion networks with conditional random fields2013
- Beyond ASR 1-best: Using word confusion networks in spoken language understandingComputer Speech & Language, 2006
- The design for the wall street journal-based CSR corpus1992