Annotated Audio Dataset of Bangladeshi English Speakers for Pause-Based Fluency Analysis

This dataset contains <b>57 audio recordings</b> of spoken English collected for the purpose of studying oral fluency, specifically through the analysis of <b>filled and silent pauses</b>. The speakers are Bangladeshi English speakers, and the recordings represent various fluency levels.The audio files are stored in the <code><strong>data/</strong></code> folder and are available in multiple formats, including <code>.mp3</code>, <code>.wav</code>, and <code>.m4a</code>.Each audio file has a corresponding <b>annotation file</b> in JSON format, located in the <code><strong>JSON/</strong></code> folder.The JSON files include detailed <b>time-stamped annotations</b> of pause markers (both filled and silent), speaker metadata, and fluency-related labels used for machine learning tasks.This dataset is intended for use in <b>speech processing, NLP, language learning research, and machine learning applications</b> related to fluency assessment. It has been used in research involving <b>transformer-based models, pause detection, and low-resource learning scenarios.</b>