Unsupervised Mandarin-Cantonese Machine Translation
This work addresses the lack of large-scale parallel data for Cantonese, a widely spoken but primarily oral language, enabling machine translation for its speakers.
The paper tackled unsupervised machine translation between Mandarin and Cantonese by creating a new corpus of about 1 million Cantonese sentences and comparing model architectures, achieving character-level BLEU scores of 25.1 and 24.4 for Mandarin-to-Cantonese and Cantonese-to-Mandarin translations, respectively.
Advancements in unsupervised machine translation have enabled the development of machine translation systems that can translate between languages for which there is not an abundance of parallel data available. We explored unsupervised machine translation between Mandarin Chinese and Cantonese. Despite the vast number of native speakers of Cantonese, there is still no large-scale corpus for the language, due to the fact that Cantonese is primarily used for oral communication. The key contributions of our project include: 1. The creation of a new corpus containing approximately 1 million Cantonese sentences, and 2. A large-scale comparison across different model architectures, tokenization schemes, and embedding structures. Our best model trained with character-based tokenization and a Transformer architecture achieved a character-level BLEU of 25.1 when translating from Mandarin to Cantonese and of 24.4 when translating from Cantonese to Mandarin. In this paper we discuss our research process, experiments, and results.