Distinguishing Repetition Disfluency from Morphological Reduplication in Bangla ASR Transcripts: A Novel Corpus and Benchmarking Analysis

Zaara Zabeen Arpa, Sadnam Sakib Apurbo, Nazia Karim Khan Oishee, Ajwad Abrar

Keywords: Morphological Reduplication, Repetition Disfluency, Bangla ASR Transcripts, Low-Resource NLP, Disfluency Detection and Correction

TL;DR: We present the first Bangla corpus distinguishing grammatical reduplication from ASR repetition errors and show fine-tuned BanglaBERT outperforms prompted LLMs on this task.

Abstract: Automatic Speech Recognition (ASR) transcripts, especially in low-resource languages like Bangla, contain a critical ambiguity: word-word repetitions can be either Repetition Disfluency (unintentional ASR error/hesitation) or Morphological Reduplication (a deliberate grammatical construct). Standard disfluency correction fails by erroneously deleting valid linguistic information. To address this, we introduce the first publicly available Bangla corpus of 20{,}000 sentences, manually annotated to explicitly distinguish between these two phenomena in noisy ASR transcripts. We benchmark this novel resource using two paradigms: state-of-the-art multilingual Large Language Models (LLMs) and task-specific fine-tuning of encoder models. LLMs achieve competitive performance (up to 82.68% accuracy) with few-shot prompting. However, fine-tuning proves superior, with the language-specific BanglaBERT model achieving the highest accuracy of 84.78% and an F1 score of 0.677. This establishes a strong, linguistically-informed baseline and provides essential data for developing sophisticated, semantic-preserving text normalization systems for Bangla.

Track: Track 2: ML Research by Muslim Authors

Email Sharing: We authorize the sharing of all author emails with Program Chairs.

Data Release: We authorize the release of our submission and author names to the public in the event of acceptance.

Non Archival Confirmation: I understand that submissions to MusIML are non-archival and can be submitted to other venues.

Submission Number: 54