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

TL;DR

This work introduces Vietnamese Multi-Dialect (ViMD) dataset, a novel comprehensive dataset capturing the rich diversity of 63 provincial dialects spoken across Vietnam, and fine-tune state-of-the-art pre-trained models for two downstream tasks: Dialect identification and Speech recognition.

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.

Figures (12)

• Figure 1: Data Collection Pipeline for the ViMD Dataset.
• Figure 2: Distribution of Audio Duration.
• Figure 3: (a) Gender-wise Distribution of ViMD, and (b) Gender unique word counts and overlap.
• Figure 4: Comparison of Duration and Number of Speakers Between Genders.
• Figure 5: Words and Unique Words Count Across Provinces.