RegSpeech12: A Regional Corpus of Bengali Spontaneous Speech Across Dialects

2025-10-28

AI Summary

Bangla exhibits rich dialectal diversity, yet suffers from a critical lack of computationally usable speech resources and dialect-aware modeling research. To address this gap, we introduce the first large-scale, spontaneously spoken Bangla corpus covering five major dialect groups and multiple regional variants, systematically annotated with phoneme-level transcriptions and fine-grained dialect labels, accompanied by comprehensive linguistic analysis. Building upon this resource, we propose a dialect-aware deep learning framework for automatic speech recognition (ASR), incorporating region-adaptive training and dialect-informed architectural design. Our models achieve significantly improved cross-dialect robustness and constitute the first empirically validated ASR system for multi-dialectal Bangla. We publicly release the full corpus, annotations, and baseline models. This work fills a fundamental void in low-resource dialectal speech technology, providing both reusable methodological insights and foundational infrastructure to advance linguistically inclusive AI.

Technology Category

• Natural Language Processing: Conversational AI/Dialog Systems
• Machine Learning: Large Multimodal Models (LMMs)
• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval

Application Category

• Search and Retrieval-Augmented AI: Multilingual and cross-lingual Web search
• User Modeling, Personalization and Recommendation: User modeling and simulation for interactive and conversational systems
• Web Mining and Content Analysis: Mining multimedia, multimodal, multilingual, cross-lingual Web data

Abstract

The Bengali language, spoken extensively across South Asia and among diasporic communities, exhibits considerable dialectal diversity shaped by geography, culture, and history. Phonological and pronunciation-based classifications broadly identify five principal dialect groups: Eastern Bengali, Manbhumi, Rangpuri, Varendri, and Rarhi. Within Bangladesh, further distinctions emerge through variation in vocabulary, syntax, and morphology, as observed in regions such as Chittagong, Sylhet, Rangpur, Rajshahi, Noakhali, and Barishal. Despite this linguistic richness, systematic research on the computational processing of Bengali dialects remains limited. This study seeks to document and analyze the phonetic and morphological properties of these dialects while exploring the feasibility of building computational models particularly Automatic Speech Recognition (ASR) systems tailored to regional varieties. Such efforts hold potential for applications in virtual assistants and broader language technologies, contributing to both the preservation of dialectal diversity and the advancement of inclusive digital tools for Bengali-speaking communities. The dataset created for this study is released for public use.

Problem

• Addressing limited computational processing research on Bengali dialects
• Documenting phonetic and morphological properties across regional varieties
• Developing Automatic Speech Recognition systems for diverse Bengali dialects

Innovation

• Building computational models for Bengali dialect analysis
• Developing Automatic Speech Recognition systems for regional varieties
• Creating a public dataset for dialect preservation and technology