English-to-Chinese Transliteration with Phonetic Back-transliteration
This work addresses the challenge of improving transliteration accuracy for languages like Chinese, Hebrew, and Thai, though it appears incremental by building on existing deep learning approaches.
The paper tackled the problem of transliterating named entities between languages by incorporating phonetic information into neural networks, achieving better or similar performance compared to state-of-the-art methods across three language pairs and six directions.
Transliteration is a task of translating named entities from a language to another, based on phonetic similarity. The task has embraced deep learning approaches in recent years, yet, most ignore the phonetic features of the involved languages. In this work, we incorporate phonetic information into neural networks in two ways: we synthesize extra data using forward and back-translation but in a phonetic manner; and we pre-train models on a phonetic task before learning transliteration. Our experiments include three language pairs and six directions, namely English to and from Chinese, Hebrew and Thai. Results indicate that our proposed approach brings benefits to the model and achieves better or similar performance when compared to state of the art.