Data Augmentation and Hyperparameter Tuning for Low-Resource MFA
This provides a workable alternative for researchers and practitioners working with endangered and under-resourced languages to improve alignment accuracy without needing large datasets.
The paper tackled the problem of lower accuracy in multilingual forced alignment for low-resource languages by comparing data augmentation and hyperparameter tuning, finding that hyperparameter tuning substantially improved performance without infeasible training time.
A continued issue for those working with computational tools and endangered and under-resourced languages is the lower accuracy of results for languages with smaller amounts of data. We attempt to ameliorate this issue by using data augmentation methods to increase corpus size, comparing augmentation to hyperparameter tuning for multilingual forced alignment. Unlike text augmentation methods, audio augmentation does not lead to substantially increased performance. Hyperparameter tuning, on the other hand, results in substantial improvement without (for this amount of data) infeasible additional training time. For languages with small to medium amounts of training data, this is a workable alternative to adapting models from high-resource languages.