Adapting Language Balance in Code-Switching Speech

TL;DR

This paper introduces a method to improve large models' performance on code-switching speech by emphasizing learning at code-switching points, reducing errors in recognizing language switches, especially in Arabic and Chinese-English datasets.

Contribution

It proposes a differentiable surrogate that highlights code-switching points during training, enhancing model robustness against context bias in code-switching scenarios.

Findings

Reduced substitution errors at code-switching points

Improved prediction accuracy for language switches

Enhanced model robustness in multilingual speech recognition

Abstract

Despite achieving impressive results on standard benchmarks, large foundational models still struggle against code-switching test cases. When data scarcity cannot be used as the usual justification for poor performance, the reason may lie in the infrequent occurrence of code-switched moments, where the embedding of the second language appears subtly. Instead of expecting the models to learn this infrequency on their own, it might be beneficial to provide the training process with labels. Evaluating model performance on code-switching data requires careful localization of code-switching points where recognition errors are most consequential, so that the analysis emphasizes mistakes occurring at those moments. Building on this observation, we leverage the difference between the embedded and the main language to highlight those code-switching points and thereby emphasize learning at those locations. This simple yet effective differentiable surrogate mitigates context bias during generation -- the central challenge in code-switching -- thereby improving the model's robustness. Our experiments with Arabic and Chinese-English showed that the models are able to predict the switching places more correctly, reflected by the reduced substitution error.