Low-resource Language Discrimination Towards Chinese Dialects with Transfer learning and Data Augmentation
Summary
The paper proposes a novel framework (CDDTLDA) using transfer learning and data augmentation to improve Chinese dialects discrimination under low-resource conditions, achieving state-of-the-art results on two benchmark corpora.
View Cached Full Text
Cached at: 06/18/26, 05:45 AM
# Low-resource Language Discrimination Towards Chinese Dialects with Transfer learning and Data Augmentation Source: [https://arxiv.org/abs/2606.18597](https://arxiv.org/abs/2606.18597) [View PDF](https://arxiv.org/pdf/2606.18597) > Abstract:Chinese dialects discrimination is a challenging natural language processing task due to scarce annotation resource\. In this article, we develop a novel Chinese dialects discrimination framework with transfer learning and data augmentation \(CDDTLDA\) in order to overcome the shortage of resources\. To be more specific, we first use a relatively larger Chinese dialects corpus to train a source\-side automatic speech recognition \(ASR\) model\. Then, we adopt a simple but effective data augmentation method \(i\.e\., speed, pitch, and noise disturbance\) to augment the target\-side low\-resource Chinese dialects, and fine\-tune another target ASR model based on the previous source\-side ASR model\. Meanwhile, the potential common semantic features between source\-side and target\-side ASR models can be captured by using self\-attention mechanism\. Finally, we extract the hidden semantic representation in the target ASR model to conduct Chinese dialects discrimination\. Our extensive experimental results demonstrate that our model significantly outperforms state\-of\-the\-art methods on two benchmark Chinese dialects corpora\. ## Submission history From: Fan Xu \[[view email](https://arxiv.org/show-email/21108fd4/2606.18597)\] **\[v1\]**Wed, 17 Jun 2026 01:46:41 UTC \(993 KB\)
Similar Articles
Speech-Driven End-to-End Language Discrimination towards Chinese Dialects
This paper investigates speech-driven features for fine-grained discrimination among Chinese dialects, using an end-to-end model that combines MFCC-based features with word-level embeddings via a CNN, outperforming text-driven methods.
DialectS2S: End-to-End Speech Dialogue Modeling for Low-Resource Chinese Dialects
DialectS2S is an end-to-end speech dialogue model for low-resource Chinese dialects, introducing a scalable data synthesis pipeline and a two-stage post-training strategy with self-aligned speech supervision. Experiments show improvements in dialect consistency, response quality, and intelligibility, with fully open-sourced models, data, and code.
Convex Low-resource Accent-Robust Language Detection in Speech Recognition
This paper introduces CLD, a lightweight convex optimization-based language detection head for ASR that achieves 97-98% accuracy with under 100 training samples while reducing compute costs by 13x, addressing accent and dialect robustness across 5 languages and 24 sub-dialects.
Multilingual Detection of Alzheimer's Disease from Speech: A Cross-Linguistic Transfer Learning Approach
This paper proposes a cross-linguistic transfer learning approach for detecting Alzheimer's Disease from speech across multiple languages, achieving F1 scores of 82% and supporting real-time screening applications.
Improving low-resource ASR using bilingual fine-tuning with language identification: a cross-linguistic evaluation
This study evaluates bilingual fine-tuning with language identification tokens for improving ASR in low-resource languages across nine diverse language pairs, finding that high LID accuracy is beneficial and that providing the LID token at inference can boost performance when LID accuracy is low.