Multi-Dialect Vietnamese: Task, Dataset, Baseline Models and Challenges

TL;DR

This paper introduces a comprehensive Vietnamese multi-dialect dataset with 63 provincial dialects, benchmarks speech recognition and dialect identification, and discusses challenges and implications for low-resource language processing.

Contribution

It presents the first fine-grained dataset of Vietnamese dialects, along with baseline models for dialect identification and speech recognition tasks.

Findings

Geographical factors influence dialect variations.

Current models face challenges with multi-dialect speech data.

The dataset enables future research in low-resource language dialects.

Abstract

Vietnamese, a low-resource language, is typically categorized into three primary dialect groups that belong to Northern, Central, and Southern Vietnam. However, each province within these regions exhibits its own distinct pronunciation variations. Despite the existence of various speech recognition datasets, none of them has provided a fine-grained classification of the 63 dialects specific to individual provinces of Vietnam. To address this gap, we introduce Vietnamese Multi-Dialect (ViMD) dataset, a novel comprehensive dataset capturing the rich diversity of 63 provincial dialects spoken across Vietnam. Our dataset comprises 102.56 hours of audio, consisting of approximately 19,000 utterances, and the associated transcripts contain over 1.2 million words. To provide benchmarks and simultaneously demonstrate the challenges of our dataset, we fine-tune state-of-the-art pre-trained models for two downstream tasks: (1) Dialect identification and (2) Speech recognition. The empirical results suggest two implications including the influence of geographical factors on dialects, and the constraints of current approaches in speech recognition tasks involving multi-dialect speech data. Our dataset is available for research purposes.