Zero-Shot End-to-End Spoken Language Understanding via Cross-Modal Selective Self-Training

TL;DR

This paper introduces a novel cross-modal selective self-training method for zero-shot end-to-end spoken language understanding, effectively addressing domain mismatch, imbalance, and noise issues to improve performance without speech-semantics pairs.

Contribution

The paper proposes CMSST, a new approach that uses clustering and a selection network to enhance zero-shot E2E SLU, along with new benchmarks for matched and mismatched speech scenarios.

Findings

CMSST outperforms previous methods in zero-shot SLU tasks.

Significant reduction in training data and time needed.

Effective handling of domain mismatch and noise issues.

Abstract

End-to-end (E2E) spoken language understanding (SLU) is constrained by the cost of collecting speech-semantics pairs, especially when label domains change. Hence, we explore \textit{zero-shot} E2E SLU, which learns E2E SLU without speech-semantics pairs, instead using only speech-text and text-semantics pairs. Previous work achieved zero-shot by pseudolabeling all speech-text transcripts with a natural language understanding (NLU) model learned on text-semantics corpora. However, this method requires the domains of speech-text and text-semantics to match, which often mismatch due to separate collections. Furthermore, using the entire collected speech-text corpus from any domains leads to \textit{imbalance} and \textit{noise} issues. To address these, we propose \textit{cross-modal selective self-training} (CMSST). CMSST tackles imbalance by clustering in a joint space of the three modalities (speech, text, and semantics) and handles label noise with a selection network. We also introduce two benchmarks for zero-shot E2E SLU, covering matched and found speech (mismatched) settings. Experiments show that CMSST improves performance in both two settings, with significantly reduced sample sizes and training time. Our code and data are released in https://github.com/amazon-science/zero-shot-E2E-slu.