Enabling ASR for Low-Resource Languages: A Comprehensive Dataset Creation Approach

TL;DR

This paper presents a new pipeline for creating ASR training datasets from audiobooks, enabling better speech recognition in low-resource languages by effectively segmenting long audio into suitable training units.

Contribution

It introduces a novel, portable method for aligning and segmenting audiobook audio for low-resource language ASR dataset creation, improving data availability and model performance.

Findings

Effective audio-text alignment and segmentation method

Application demonstrated on Armenian language

Enhanced ASR performance for low-resource languages

Abstract

In recent years, automatic speech recognition (ASR) systems have significantly improved, especially in languages with a vast amount of transcribed speech data. However, ASR systems tend to perform poorly for low-resource languages with fewer resources, such as minority and regional languages. This study introduces a novel pipeline designed to generate ASR training datasets from audiobooks, which typically feature a single transcript associated with hours-long audios. The common structure of these audiobooks poses a unique challenge due to the extensive length of audio segments, whereas optimal ASR training requires segments ranging from 4 to 15 seconds. To address this, we propose a method for effectively aligning audio with its corresponding text and segmenting it into lengths suitable for ASR training. Our approach simplifies data preparation for ASR systems in low-resource languages and demonstrates its application through a case study involving the Armenian language. Our method, which is "portable" to many low-resource languages, not only mitigates the issue of data scarcity but also enhances the performance of ASR models for underrepresented languages.