Research

Paper

TESTING March 05, 2026

PersianPunc: A Large-Scale Dataset and BERT-Based Approach for Persian Punctuation Restoration

Authors

Mohammad Javad Ranjbar Kalahroodi, Heshaam Faili, Azadeh Shakery

Abstract

Punctuation restoration is essential for improving the readability and downstream utility of automatic speech recognition (ASR) outputs, yet remains underexplored for Persian despite its importance. We introduce PersianPunc, a large-scale, high-quality dataset of 17 million samples for Persian punctuation restoration, constructed through systematic aggregation and filtering of existing textual resources. We formulate punctuation restoration as a token-level sequence labeling task and fine-tune ParsBERT to achieve strong performance. Through comparative evaluation, we demonstrate that while large language models can perform punctuation restoration, they suffer from critical limitations: over-correction tendencies that introduce undesired edits beyond punctuation insertion (particularly problematic for speech-to-text pipelines) and substantially higher computational requirements. Our lightweight BERT-based approach achieves a macro-averaged F1 score of 91.33% on our test set while maintaining efficiency suitable for real-time applications. We make our dataset (https://huggingface.co/datasets/MohammadJRanjbar/persian-punctuation-restoration) and model (https://huggingface.co/MohammadJRanjbar/parsbert-persian-punctuation) publicly available to facilitate future research in Persian NLP and provide a scalable framework applicable to other morphologically rich, low-resource languages.

Metadata

arXiv ID: 2603.05314
Provider: ARXIV
Primary Category: cs.CL
Published: 2026-03-05
Fetched: 2026-03-06 14:20

Related papers

Raw Data (Debug)
{
  "raw_xml": "<entry>\n    <id>http://arxiv.org/abs/2603.05314v1</id>\n    <title>PersianPunc: A Large-Scale Dataset and BERT-Based Approach for Persian Punctuation Restoration</title>\n    <updated>2026-03-05T15:55:55Z</updated>\n    <link href='https://arxiv.org/abs/2603.05314v1' rel='alternate' type='text/html'/>\n    <link href='https://arxiv.org/pdf/2603.05314v1' rel='related' title='pdf' type='application/pdf'/>\n    <summary>Punctuation restoration is essential for improving the readability and downstream utility of automatic speech recognition (ASR) outputs, yet remains underexplored for Persian despite its importance. We introduce PersianPunc, a large-scale, high-quality dataset of 17 million samples for Persian punctuation restoration, constructed through systematic aggregation and filtering of existing textual resources. We formulate punctuation restoration as a token-level sequence labeling task and fine-tune ParsBERT to achieve strong performance. Through comparative evaluation, we demonstrate that while large language models can perform punctuation restoration, they suffer from critical limitations: over-correction tendencies that introduce undesired edits beyond punctuation insertion (particularly problematic for speech-to-text pipelines) and substantially higher computational requirements. Our lightweight BERT-based approach achieves a macro-averaged F1 score of 91.33% on our test set while maintaining efficiency suitable for real-time applications. We make our dataset (https://huggingface.co/datasets/MohammadJRanjbar/persian-punctuation-restoration) and model (https://huggingface.co/MohammadJRanjbar/parsbert-persian-punctuation) publicly available to facilitate future research in Persian NLP and provide a scalable framework applicable to other morphologically rich, low-resource languages.</summary>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.CL'/>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.AI'/>\n    <published>2026-03-05T15:55:55Z</published>\n    <arxiv:primary_category term='cs.CL'/>\n    <author>\n      <name>Mohammad Javad Ranjbar Kalahroodi</name>\n    </author>\n    <author>\n      <name>Heshaam Faili</name>\n    </author>\n    <author>\n      <name>Azadeh Shakery</name>\n    </author>\n  </entry>"
}