Modeling Homophone Noise for Robust Neural Machine Translation

2020/12/15 by Qin, Wenjie, Li, Xiang, Sun, Yuhui +3
#Computation and Language (cs.CL) #FOS: Computer and information sciences

paper · doi:10.48550/arxiv.2012.08396

Abstract

In this paper, we propose a robust neural machine translation (NMT) framework. The framework consists of a homophone noise detector and a syllable-aware NMT model to homophone errors. The detector identifies potential homophone errors in a textual sentence and converts them into syllables to form a mixed sequence that is then fed into the syllable-aware NMT. Extensive experiments on Chinese->English translation demonstrate that our proposed method not only significantly outperforms baselines on noisy test sets with homophone noise, but also achieves a substantial improvement on clean text.

Related

• Robust Neural Machine Translation with Joint Textual and Phonetic Embedding
• Improving the Robustness of Speech Translation
• Synthetic and Natural Noise Both Break Neural Machine Translation
• Addressing the Vulnerability of NMT in Input Perturbations
• Robust Neural Machine Translation: Modeling Orthographic and\n Interpunctual Variation