Code-Switched Language Models Using Neural Based Synthetic Data from Parallel Sentences
This addresses data scarcity for code-switching in NLP, particularly for distant languages, though it is incremental as it builds on prior synthetic data generation methods.
The paper tackles the problem of training code-switched language models by proposing a neural sequence-to-sequence model with a copy mechanism to generate synthetic data from parallel sentences, achieving state-of-the-art performance and improving automatic speech recognition.
Training code-switched language models is difficult due to lack of data and complexity in the grammatical structure. Linguistic constraint theories have been used for decades to generate artificial code-switching sentences to cope with this issue. However, this require external word alignments or constituency parsers that create erroneous results on distant languages. We propose a sequence-to-sequence model using a copy mechanism to generate code-switching data by leveraging parallel monolingual translations from a limited source of code-switching data. The model learns how to combine words from parallel sentences and identifies when to switch one language to the other. Moreover, it captures code-switching constraints by attending and aligning the words in inputs, without requiring any external knowledge. Based on experimental results, the language model trained with the generated sentences achieves state-of-the-art performance and improves end-to-end automatic speech recognition.