Meta-Transfer Learning for Code-Switched Speech Recognition
This addresses the challenge of limited data for code-switched speech recognition, which affects multilingual speakers, but it is incremental as it builds on transfer learning methods.
The paper tackles the problem of building speech recognition systems for code-switched (mixed-language) speech in low-resource settings by proposing meta-transfer learning, which extracts information from high-resource monolingual datasets to improve recognition, resulting in outperforming existing baselines and faster convergence.
An increasing number of people in the world today speak a mixed-language as a result of being multilingual. However, building a speech recognition system for code-switching remains difficult due to the availability of limited resources and the expense and significant effort required to collect mixed-language data. We therefore propose a new learning method, meta-transfer learning, to transfer learn on a code-switched speech recognition system in a low-resource setting by judiciously extracting information from high-resource monolingual datasets. Our model learns to recognize individual languages, and transfer them so as to better recognize mixed-language speech by conditioning the optimization on the code-switching data. Based on experimental results, our model outperforms existing baselines on speech recognition and language modeling tasks, and is faster to converge.