Synthetic Source Language Augmentation for Colloquial Neural Machine Translation
This work tackles the problem of translating colloquial Indonesian for NMT models, which is an incremental improvement for the NMT community.
This paper addresses the challenge of colloquial language in Neural Machine Translation (NMT) by creating a new Indonesian-English test set from YouTube transcripts and Twitter. They show that synthetically augmenting formal Indonesian with colloquial style improves baseline Id-En models on this new test data, as measured by BLEU.
Neural machine translation (NMT) is typically domain-dependent and style-dependent, and it requires lots of training data. State-of-the-art NMT models often fall short in handling colloquial variations of its source language and the lack of parallel data in this regard is a challenging hurdle in systematically improving the existing models. In this work, we develop a novel colloquial Indonesian-English test-set collected from YouTube transcript and Twitter. We perform synthetic style augmentation to the source of formal Indonesian language and show that it improves the baseline Id-En models (in BLEU) over the new test data.