Mining Word Boundaries in Speech as Naturally Annotated Word Segmentation Data

TL;DR

This paper introduces a novel method to mine word boundaries from parallel speech and text data, leveraging character alignments and pause cues, to improve Chinese word segmentation especially in low-resource settings.

Contribution

It is the first to mine word boundaries from parallel speech/text data for Chinese word segmentation, utilizing a complete-then-train strategy to enhance model performance.

Findings

Significant improvement in CWS accuracy in low-resource scenarios

Effective use of naturally annotated speech/text data for segmentation

Demonstrated cross-domain robustness

Abstract

Inspired by early research on exploring naturally annotated data for Chinese word segmentation (CWS), and also by recent research on integration of speech and text processing, this work for the first time proposes to mine word boundaries from parallel speech/text data. First we collect parallel speech/text data from two Internet sources that are related with CWS data used in our experiments. Then, we obtain character-level alignments and design simple heuristic rules for determining word boundaries according to pause duration between adjacent characters. Finally, we present an effective complete-then-train strategy that can better utilize extra naturally annotated data for model training. Experiments demonstrate our approach can significantly boost CWS performance in both cross-domain and low-resource scenarios.