Robust Neural Machine Translation with Joint Textual and Phonetic Embedding
arXiv:1810.06729
Abstract
Neural machine translation (NMT) is notoriously sensitive to noises, but noises are almost inevitable in practice. One special kind of noise is the homophone noise, where words are replaced by other words with similar pronunciations. We propose to improve the robustness of NMT to homophone noises by 1) jointly embedding both textual and phonetic information of source sentences, and 2) augmenting the training dataset with homophone noises. Interestingly, to achieve better translation quality and more robustness, we found that most (though not all) weights should be put on the phonetic rather than textual information. Experiments show that our method not only significantly improves the robustness of NMT to homophone noises, but also surprisingly improves the translation quality on some clean test sets.
arXiv commentACL 2019
References in corpus (8)
- Sequence to Sequence Learning with Neural Networks
- Intriguing properties of neural networks
- Learning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translation
- Convolutional Sequence to Sequence Learning
- Synthetic and Natural Noise Both Break Neural Machine Translation
- Assessing the Tolerance of Neural Machine Translation Systems Against Speech Recognition Errors
- Improving the Robustness of Speech Translation
- STACL: Simultaneous Translation with Implicit Anticipation and Controllable Latency using Prefix-to-Prefix Framework