ML-LMCL: Mutual Learning and Large-Margin Contrastive Learning for Improving ASR Robustness in Spoken Language Understanding

Abstract

Spoken language understanding (SLU) is a fundamental task in the task-oriented dialogue systems. However, the inevitable errors from automatic speech recognition (ASR) usually impair the understanding performance and lead to error propagation. Although there are some attempts to address this problem through contrastive learning, they (1) treat clean manual transcripts and ASR transcripts equally without discrimination in fine-tuning; (2) neglect the fact that the semantically similar pairs are still pushed away when applying contrastive learning; (3) suffer from the problem of Kullback-Leibler (KL) vanishing. In this paper, we propose Mutual Learning and Large-Margin Contrastive Learning (ML-LMCL), a novel framework for improving ASR robustness in SLU. Specifically, in fine-tuning, we apply mutual learning and train two SLU models on the manual transcripts and the ASR transcripts, respectively, aiming to iteratively share knowledge between these two models. We also introduce a distance polarization regularizer to avoid pushing away the intra-cluster pairs as much as possible. Moreover, we use a cyclical annealing schedule to mitigate KL vanishing issue. Experiments on three datasets show that ML-LMCL outperforms existing models and achieves new state-of-the-art performance.

Towards Explainable Joint Models via Information Theory for Multiple Intent Detection and Slot Filling

“…) to achieve the synchronous optimization. Thirdly, inspired by Ji et al (2022b); Cheng et al (2023a), we utilize contrastive learning to minimize the In-foNCE loss to maximize a lower bound on CIG.…”

The Interpreter Understands Your Meaning: End-to-end Spoken Language Understanding Aided by Speech Translation

“…End-to-end SLU Cascaded SLU methods work on ASR transcripts, for which error propagation is a major challenge (Chang and Chen, 2022;Cheng et al, 2023a). Hence recently end-to-end methods have gained popularity (Serdyuk et al, 2018;Haghani et al, 2018), especially with the performance gap compared with cascaded systems mitigated in many cases thanks to the PTLM paradigm.…”