Improving Spoken Language Identification with Map-Mix

TL;DR

This paper introduces Map-Mix, a data augmentation technique leveraging model training dynamics to improve dialect classification in low-resource settings, enhancing model calibration and performance.

Contribution

The paper presents a novel datamaps-based mixup method that improves dialect identification accuracy and calibration in low-resource scenarios.

Findings

Map-Mix improves weighted F1 scores by 2% over baseline.

The method results in a more well-calibrated model.

Effective in low-resource dialect classification tasks.

Abstract

The pre-trained multi-lingual XLSR model generalizes well for language identification after fine-tuning on unseen languages. However, the performance significantly degrades when the languages are not very distinct from each other, for example, in the case of dialects. Low resource dialect classification remains a challenging problem to solve. We present a new data augmentation method that leverages model training dynamics of individual data points to improve sampling for latent mixup. The method works well in low-resource settings where generalization is paramount. Our datamaps-based mixup technique, which we call Map-Mix improves weighted F1 scores by 2% compared to the random mixup baseline and results in a significantly well-calibrated model. The code for our method is open sourced on https://github.com/skit-ai/Map-Mix.