---
title: "Meaning Over Morphology — Niloy 2026"
description: "Bangla regional dialects such as Sylheti and Chittagonian differ from standard Bangla so much that they can approach mutual unintelligibility."
authors: ["Soumik Deb Niloy", "Subhey Sadi Rahman", "Mahbub E Sobhani", "Md. Golam Rabiul Alam", "Farig Yousuf Sadeque", "Md. Rezuwan Hassan"]
year: 2026
venue: "Proceedings of the First Workshop on Dialects in NLP –- A Resource Perspective"
source: "https://aclanthology.org/2026.dialres-1.24/"
doi: "10.63317/3h8y7fs2zbq4"
licence: "CC BY-NC 4.0 (ELRA, LREC proceedings)"
licence_url: "https://creativecommons.org/licenses/by-nc/4.0/"
summary_licence: "Elamur Cosmic License"
canonical: "https://research.elamur.ai/free/papers/niloy-2026-meaning-over-morphology-multi-metric/"
---
[Elamur Research](https://md.research.elamur.ai/) · [this page in HTML](https://research.elamur.ai/free/papers/niloy-2026-meaning-over-morphology-multi-metric/)

[home](https://md.research.elamur.ai/) / [papers](https://md.research.elamur.ai/free/papers/) / [2026](https://md.research.elamur.ai/free/papers/2026/page/3/) / Meaning Over Morphology: A Multi-Metric Benchmark of LLMs for Bangla Dialect Translation

# Meaning Over Morphology: A Multi-Metric Benchmark of LLMs for Bangla Dialect Translation

**Authors:** Soumik Deb Niloy, Subhey Sadi Rahman, Mahbub E Sobhani, Md. Golam Rabiul Alam, Farig Yousuf Sadeque, Md. Rezuwan Hassan · **Year:** 2026 · **Venue:** Proceedings of the First Workshop on Dialects in NLP –- A Resource Perspective · **Licence:** [CC BY-NC 4.0 (ELRA, LREC proceedings)](https://creativecommons.org/licenses/by-nc/4.0/)

**Links:** [Publisher's page](https://aclanthology.org/2026.dialres-1.24/) · [PDF](https://aclanthology.org/2026.dialres-1.24.pdf) · [DOI 10.63317/3h8y7fs2zbq4](https://doi.org/10.63317/3h8y7fs2zbq4) · [OpenAlex](https://openalex.org/W7166549334)

**Citations:** 0 citations (openalex-snapshot · 2026-10-14)

## Summary
Bangla regional dialects such as Sylheti and Chittagonian differ from standard Bangla so much that they can approach mutual unintelligibility. Because foundation models learn mostly from standard text, dialect speakers are poorly served. The authors benchmark eight open-source LLMs on converting fifteen Bangla dialects into Standard Bangla. For this they assemble the Regional-Standard Bangla Corpus (RSBC), which merges nine public parallel datasets with a new, still-private subset. The merge involved schema normalization, UTF-8 standardization, exact and fuzzy deduplication, and quality checks. Outputs are scored with surface-level measures (BLEU, WER, CER) and with semantics-aware ones (METEOR, COMET, BERTScore), followed by a per-dialect linguistic error analysis. Semantic metrics come out consistently higher than exact-match ones. This suggests that models often keep the meaning while choosing different word forms, which n-gram scores penalize. The work offers a reproducible evaluation setup and argues that dialect MT should be judged on meaning, not only surface overlap. That point carries over to other low-resource dialect settings.

## Key points
- **RSBC corpus:** 68,666 parallel dialect–standard pairs. Of these, 51,662 come from external sources, including Vashantor, BIDWESH, Kothon, BanglaDialecto, ONUBAD, FeniVerse and ChatgaiyyaAlap. The other 17,004 come from the Proprietary Dialect Dataset (PDD).
- **PDD:** covers 8 regions. Native-speaker annotators (9 per dialect) worked for about six months, with ongoing validation by linguists. The subset stays closed until its parent project is finished.
- **Dialect coverage:** 15 districts, very unevenly sized. Chittagong is the largest with 19,919 samples, and Tangail is the smallest with 350.
- **Models:** gpt-oss:20B, Qwen3:8B, Phi4:14B, Granite4:3B, deepseek-r1:7B, Gemma2:9B, LLaMA3.2:3B and Mistral:7B.
- **Score patterns:** BLEU stays low, mostly around 0.05–0.35, while BERTScore mostly falls between about 0.6 and 0.89. WER sometimes exceeds 1.0 for the weaker models. In the district tables we received, gpt-oss:20B usually ranks near the top and LLaMA3.2:3B and Mistral:7B near the bottom. Part of the results section was missing from our copy, so these rankings should be checked against the full paper.

*Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.*

Fields: [Machine translation](https://md.research.elamur.ai/free/shelves/machine-translation/) · [Natural language processing](https://md.research.elamur.ai/free/shelves/nlp/)

## Cite

The paper itself, not this page. BibTeX from the library's record of the work.

```bibtex
@inproceedings{niloy2026meaning,
  title = {{Meaning Over Morphology: A Multi-Metric Benchmark of LLMs for Bangla Dialect Translation}},
  author = {Niloy, Soumik Deb and Rahman, Subhey Sadi and Sobhani, Mahbub E and Alam, Md. Golam Rabiul and Sadeque, Farig Yousuf and Hassan, Md. Rezuwan},
  booktitle = {Proceedings of the First Workshop on Dialects in NLP –- A Resource Perspective},
  year = {2026},
  doi = {10.63317/3h8y7fs2zbq4},
  url = {https://doi.org/10.63317/3h8y7fs2zbq4}
}
```

- [← previous Meaning Annotation Experience. A Tribute to Petr Sgall](https://md.research.elamur.ai/free/papers/mikulova-2026-meaning-annotation-experience-tribute-petr/)
- [next → Meaning-Making Process and Error Dynamics in ChatGPT-Mediated Translation](https://md.research.elamur.ai/free/papers/mihalache-2026-meaning-making-process-error-dynamics/)
- [↑ back to papers of 2026](https://md.research.elamur.ai/free/papers/2026/page/3/)

## References

30 references, 5 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Lee, Yin Tat Lee, Yuanzhi Li, Weishung Liu, Caio C. T. Mendes, Anh Nguyen, Eric Price, Gustavo de Rosa, Olli Saarikivi, Adil Salim, Shital Shah, Xin Wang, Rachel Ward, Yue Wu, Dingli Yu, Cyril Zhang, and Yi Zhang. 2024. Phi-4 technical report. [find](https://scholar.google.com/scholar?q=Lee%2C+Yin+Tat+Lee%2C+Yuanzhi+Li%2C+Weishung+Liu%2C+Caio+C%2E+T%2E+Mendes%2C+Anh+Nguyen%2C+Eric+Price%2C+Gustavo+de+Rosa%2C+Olli+Saarikivi%2C+Adil+Salim%2C+Shital+Shah%2C+Xin+Wang%2C+Rachel+Ward%2C+Yue+Wu%2C+Dingli+Yu%2C+Cyril+Zhang%2C+) [our card](https://md.research.elamur.ai/free/papers/abdin-2024-phi-4-technical-report/)
2. Sandhini Agarwal, Lama Ahmad, Jason Ai, Sam Altman, Andy Applebaum, Edwin Arbus, Rahul K. Arora, Yu Bai, Bowen Baker, Haiming Bao, et al. 2025. gpt-oss-120b & gpt-oss-20b model card. Md. Tofael Ahmed Bhuiyan, Md. [find](https://scholar.google.com/scholar?q=Sandhini+Agarwal%2C+Lama+Ahmad%2C+Jason+Ai%2C+Sam+Altman%2C+Andy+Applebaum%2C+Edwin+Arbus%2C+Rahul+K%2E+Arora%2C+Yu+Bai%2C+Bowen+Baker%2C+Haiming+Bao%2C+et+al%2E+2025%2E+gpt%2Doss%2D120b+%26+gpt%2Doss%2D20b+model+card%2E+Md%2E+Tofael+Ahmed+) [our card](https://md.research.elamur.ai/free/papers/agarwal-2025-gpt-oss-120b-gpt-oss/)
3. Abdur Rahman, and Abdul Kadar Muhammad Masum. 2025. BhasaBodh: Bridging Bangla dialects and Romanized forms through machine translation. In Proceedings of the Second Workshop on Bangla Language Processing (BLP-2025), pages 113–118, Mumbai, India. Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Abdur+Rahman%2C+and+Abdul+Kadar+Muhammad+Masum%2E+2025%2E+BhasaBodh%3A+Bridging+Bangla+dialects+and+Romanized+forms+through+machine+translation%2E+In+Proceedings+of+the+Second+Workshop+on+Bangla+Language+Proces)
4. DeepSeek-AI. 2025. Deepseek-r1: Incentivizing reasoning capability in llms via reinforcement learning. Tawsif Tashwar Dipto, Azmol Hossain, Rubayet Sabbir Faruque, Md. Rezuwan Hassan, Kanij Fatema, Tanmoy Shome, Ruwad Naswan, Md.Foriduzzaman Zihad, Mohaymen Ul Anam, Nazia Tasnim, Hasan Mahmud, Md Kamrul Hasan, Md. [find](https://scholar.google.com/scholar?q=DeepSeek%2DAI%2E+2025%2E+Deepseek%2Dr1%3A+Incentivizing+reasoning+capability+in+llms+via+reinforcement+learning%2E+Tawsif+Tashwar+Dipto%2C+Azmol+Hossain%2C+Rubayet+Sabbir+Faruque%2C+Md%2E+Rezuwan+Hassan%2C+Kanij+Fatema%2C+Ta) [our card](https://md.research.elamur.ai/free/papers/deepseek-2025-ai-deepseek-r1-incentivizing-reasoning-capability/)
5. Mehedi Hasan Shawon, Farig Sadeque, and Tahsin Reasat. 2025. Are ASR foundation models generalized enough to capture features of regional dialects for low-resource languages? In Proceedings of the 14th International Joint Conference on Natural Language Processing and the 4th Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics, pages 178–188, Mumbai, India. The Asian Federation of Natural Language Processing and The Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Mehedi+Hasan+Shawon%2C+Farig+Sadeque%2C+and+Tahsin+Reasat%2E+2025%2E+Are+ASR+foundation+models+generalized+enough+to+capture+features+of+regional+dialects+for+low%2Dresource+languages%3F+In+Proceedings+of+the+14t)
6. Aaron Grattafiori, Abhimanyu Dubey, Abhinav Jauhri, Abhinav Pandey, Abhishek Kadian, Ahmad Al-Dahle, Aiesha Letman, Akhil Mathur, Alan Schelten, Amy Yang, Angela Fan, Anirudh Goyal, Anthony Hartshorn, and …. 2024. The llama 3 herd of models. arXiv preprint, 2407.21783. Md. Rezuwan Hassan, Azmol Hossain, Kanij Fatema, Rubayet Sabbir Faruque, Tanmoy Shome, Ruwad Naswan, Trina Chakraborty, Md. Foriduzzaman Zihad, Tawsif Tashwar Dipto, Nazia Tasnim, Nazmuddoha Ansary, Md. Mehedi Hasan Shawon, Ahmed Imtiaz Humayun, Md. [find](https://scholar.google.com/scholar?q=Aaron+Grattafiori%2C+Abhimanyu+Dubey%2C+Abhinav+Jauhri%2C+Abhinav+Pandey%2C+Abhishek+Kadian%2C+Ahmad+Al%2DDahle%2C+Aiesha+Letman%2C+Akhil+Mathur%2C+Alan+Schelten%2C+Amy+Yang%2C+Angela+Fan%2C+Anirudh+Goyal%2C+Anthony+Hartshorn%2C)
7. Golam Rabiul Alam, Farig Sadeque, and Asif Sushmit. 2025. Regspeech12: A regional corpus of bengali spontaneous speech across dialects. [find](https://scholar.google.com/scholar?q=Golam+Rabiul+Alam%2C+Farig+Sadeque%2C+and+Asif+Sushmit%2E+2025%2E+Regspeech12%3A+A+regional+corpus+of+bengali+spontaneous+speech+across+dialects%2E)
8. Albert Q. Jiang, Alexandre Sablayrolles, Arthur Mensch, Chris Bamford, Devendra Singh Chaplot, Diego de las Casas, Florian Bressand, Gianna Lengyel, Guillaume Lample, Lucile Saulnier, Lélio Renard Lavaud, Marie-Anne Lachaux, Pierre Stock, Teven Le Scao, Thibaut Lavril, Thomas Wang, Timothée Lacroix, and William El Sayed. 2023. Mistral 7b. arXiv preprint arXiv:2310.06825. Md. [↗](https://arxiv.org/abs/2310.06825)
9. Arafat Alam Khandaker, Ziyan Shirin Raha, Bidyarthi Paul, and Tashreef Muhammad. 2025. Bridging dialects: Translating standard bangla to regional variants using neural models. [find](https://scholar.google.com/scholar?q=Arafat+Alam+Khandaker%2C+Ziyan+Shirin+Raha%2C+Bidyarthi+Paul%2C+and+Tashreef+Muhammad%2E+2025%2E+Bridging+dialects%3A+Translating+standard+bangla+to+regional+variants+using+neural+models%2E)
10. Hui-Chi Kuo and Yun-Nung Chen. 2023. Zero-shot prompting for implicit intent prediction and recommendation with commonsense reasoning. In Findings of the Association for Computational Linguistics: ACL 2023, pages 249–258, Toronto, Canada. Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Hui%2DChi+Kuo+and+Yun%2DNung+Chen%2E+2023%2E+Zero%2Dshot+prompting+for+implicit+intent+prediction+and+recommendation+with+commonsense+reasoning%2E+In+Findings+of+the+Association+for+Computational+Linguistics%3A+ACL)
11. Alon Lavie and Abhaya Agarwal. 2007. ME-TEOR: An automatic metric for MT evaluation with high levels of correlation with human judgments. In Proceedings of the Second Workshop on Statistical Machine Translation, pages 228– 231, Prague, Czech Republic. Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Alon+Lavie+and+Abhaya+Agarwal%2E+2007%2E+ME%2DTEOR%3A+An+automatic+metric+for+MT+evaluation+with+high+levels+of+correlation+with+human+judgments%2E+In+Proceedings+of+the+Second+Workshop+on+Statistical+Machine+T)
12. Mayank Mishra, Matt Stallone, Gaoyuan Zhang, Yikang Shen, Aditya Prasad, Adriana Meza Soria, Michele Merler, Parameswaran Selvam, Saptha Surendran, Shivdeep Singh, Manish Sethi, Xuan-Hong Dang, Pengyuan Li, Kun-Lung Wu, Syed Zawad, Andrew Coleman, Matthew White, Mark Lewis, Raju Pavuluri, Yan Koyfman, Boris Lublinsky, Maximilien de Bayser, Ibrahim Abdelaziz, Kinjal Basu, Mayank Agarwal, Yi Zhou, Chris Johnson, Aanchal Goyal, Hima Patel, Yousaf Shah, Petros Zerfos, Heiko Ludwig, Asim Munawar, Maxwell Crouse, Pavan Kapanipathi, Shweta Salaria, Bob Calio, Sophia Wen, Seetharami Seelam, Brian Belgodere, Carlos Fonseca, Amith Singhee, Nirmit Desai, David D. Cox, Ruchir Puri, and Rameswar Panda. 2024. Granite code models: A family of open foundation models for code intelligence. arXiv preprint arXiv:2405.04324. [↗](https://arxiv.org/abs/2405.04324)
13. Kishore Papineni, Salim Roukos, Todd Ward, and Wei-Jing Zhu. 2002. Bleu: a method for automatic evaluation of machine translation. In Proceedings of the 40th Annual Meeting of the Association for Computational Linguistics, pages 311–318, Philadelphia, Pennsylvania, USA. Association for Computational Linguistics. [↗](https://doi.org/10.3115/1073083.1073135) [our card](https://md.research.elamur.ai/free/papers/papineni-2002-bleu-method-automatic-evaluation-machine/)
14. Tabia Tanzin Prama. 2025. LLMs for low-resource dialect translation using context-aware prompting: A case study on Sylheti. In Proceedings of the Second Workshop on Bangla Language Processing (BLP-2025), pages 292–308, Mumbai, India. Association for Computational Linguistics. [↗](https://doi.org/10.48448/rde3-g876)
15. Ricardo Rei, Craig Stewart, Ana C Farinha, and Alon Lavie. 2020. COMET: A neural framework for MT evaluation. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 2685– 2702, Online. Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Ricardo+Rei%2C+Craig+Stewart%2C+Ana+C+Farinha%2C+and+Alon+Lavie%2E+2020%2E+COMET%3A+A+neural+framework+for+MT+evaluation%2E+In+Proceedings+of+the+2020+Conference+on+Empirical+Methods+in+Natural+Language+Processing+) [our card](https://md.research.elamur.ai/free/papers/rei-2020-comet-neural-framework-mt-evaluation/)
16. Nils Reimers and Iryna Gurevych. 2019. Sentence-bert: Sentence embeddings using siamese bert-networks. [find](https://scholar.google.com/scholar?q=Nils+Reimers+and+Iryna+Gurevych%2E+2019%2E+Sentence%2Dbert%3A+Sentence+embeddings+using+siamese+bert%2Dnetworks%2E)
17. Md Jahid Alam Riad, Prosenjit Roy, Mahfuzur Rahman Shuvo, Nobanul Hasan, Stabak Das, Fateha Jannat Ayrin, Syeda Sadia Alam, Afsana Khan, Md Tanzim Reza, and Md Mizanur Rahman. 2025. Fine-tuning large language models for regional dialect comprehended question answering in bangla. In 2025 IEEE International Students’ Conference on Electrical, Electronics and Computer Science (SCEECS), pages 1–6. [find](https://scholar.google.com/scholar?q=Md+Jahid+Alam+Riad%2C+Prosenjit+Roy%2C+Mahfuzur+Rahman+Shuvo%2C+Nobanul+Hasan%2C+Stabak+Das%2C+Fateha+Jannat+Ayrin%2C+Syeda+Sadia+Alam%2C+Afsana+Khan%2C+Md+Tanzim+Reza%2C+and+Md+Mizanur+Rahman%2E+2025%2E+Fine%2Dtuning+large+)
18. Gemma Team, Morgane Riviere, Shreya Pathak, Pier Giuseppe Sessa, Cassidy Hardin, Surya Bhupatiraju, Léonard Hussenot, Thomas Mesnard, Bobak Shahriari, Alexandre Ramé, Johan Ferret, Peter Liu, Pouya Tafti, Abe Friesen, Michelle Casbon, Sabela Ramos, Ravin Kumar, Charline Le Lan, Sammy Jerome, Anton Tsitsulin, Nino Vieillard, Piotr Stanczyk, and (and many others). 2024. Gemma 2: Improving open language models at a practical size. CoRR, abs/2408.00118. An Yang, Anfeng Li, Baosong Yang, Beichen Zhang, Binyuan Hui, Bo Zheng, Bowen Yu, Chang Gao, Chengen Huang, Chenxu Lv, Chujie Zheng, Dayiheng Liu, Fan Zhou, Fei Huang, Feng Hu, Hao Ge, Haoran Wei, Huan Lin, Jialong Tang, Jian Yang, Jianhong Tu, Jianwei Zhang, and Qwen3 technical report. [↗](https://arxiv.org/abs/2408.00118)
19. Tianyi Zhang, Varsha Kishore, Felix Wu, Kilian Q. Weinberger, and Yoav Artzi. 2020. Bertscore: Evaluating text generation with bert. 11. [↗](https://doi.org/10.48550/arxiv.1904.09675)
20. Language Resource References Ahmed, Md Tofael and Koli, Zannatul Mawa and Rahat, Azmain Mahtab and Akter, Taslima and Ayman, Umme. 2026. BanglaRe-gionalTextCorpus: A curated dataset for four regional bangla dialects with standard Bangla and English translation. PID https://doi.org/10.17632/92r62h4k5k.4. [↗](https://doi.org/10.17632/92r62h4k5k.4)
21. Chatterji, Suniti Kumar. 1970. The Origin and Development of the Bengali Language: Volume One. Routledge, 1. PID https://doi.org/10.4324/9781003480945. [↗](https://doi.org/10.4324/9781003480945)
22. Chowdhury, Sinthia and Remal, Deawan Rakin Ahamed and Pasha, Syed Tangim and Islam, Ashraful and Noori, Sheak ChatgaiyyaAlap: A Rashed Haider. 2025. dataset for conversion from Chittagonian dialect to standard Bangla. Elsevier. PID https://doi.org/10.17632/wtms9xbkkw.1. Faisal, Md. [↗](https://doi.org/10.17632/wtms9xbkkw.1)
23. Atique and Sadaf, Farhan and Chowdhury, Dipta and Azrof, H.M. and Tanmay, Monojit Paul. 2025. Kothon: A Large-Scale Dataset for Machine Translation of the Chittagonian and Sylheti Dialects into Standard Bangla. Mendeley Data. PID https://doi.org/10.17632/2fv6vf9v2z.3. Version 3. Faria, Fatema Tuj Johora and Moin, Mukaffi Bin and Wase, Ahmed Al and Ahmmed, Mehidi and Sani, Md. [↗](https://doi.org/10.17632/2fv6vf9v2z.3)
24. Rabius and Muhammad, Tashreef. 2023. Vashantor: A Large-scale Multilingual Benchmark Dataset for Automated Translation of Bangla Regional Dialects to Bangla Language. Mendeley Data. PID https://doi.org/10.17632/bj5jgk878b.2. ArXiv preprint arXiv:2311.11142. Fayaz, Azizul Hakim and Uddin, MD. Shorif and Bhuiyan, Rayhan Uddin and Sultana, Zakia and Islam, Md. [↗](https://doi.org/10.17632/bj5jgk878b.2)
25. Samiul and Paul, Bidyarthi and Muhammad, Tashreef and Manzoor, Shahriar. 2025. BIDWESH: A Bangla Regional Based Hate Speech Detection Dataset. Mendeley Data. PID https://doi.org/10.17632/bpkrvf882k.1. ArXiv preprint arXiv:2507.16183. [↗](https://doi.org/10.17632/bpkrvf882k.1)
26. Mahi, Mehraj Hossain and Khan, Anzir Rahman and Hoque, Zesanul and Mojumdar, Mayen Uddin. 2025a. FeniVerse: A parallel corpus of Feni dialect, standard Bengali, and English. Elsevier. PID https://doi.org/10.17632/25r923frfj.1. [↗](https://doi.org/10.17632/25r923frfj.1)
27. Mahi, Mehraj Hossain and Khan, Anzir Rahman and Mojumdar, Mayen Uddin. 2025b. BanglaDial: A merged and imbalanced text dataset for Bengali regional dialect analysis. Elsevier. PID https://doi.org/10.17632/sm63ryv5dt.1. Naeen, Md. [↗](https://doi.org/10.17632/sm63ryv5dt.1)
28. Julkar and Das, Sourav Kumar and Supta, Supta Das Dip and Alahi, Tabib E and Faisal, Md Faisal Tajwar and Rahat, Abdullah Al and Ayshe, Samurtha Jahan and Mahjabeen, Ummay and Tahmid, Md Tanvir and Mojumdar, Mayen Uddin. 2025. Bangla Dialect Dataset: Exploring Linguistic Diversity Across Regions. Mendeley Data. PID https://doi.org/10.17632/sm63ryv5dt. Samin, Md. [↗](https://doi.org/10.17632/sm63ryv5dt)
29. Nazmus Sadat and Ahad, Jawad Ibn and Medha, Tanjila Ahmed and Rahman, Fuad and Amin, Mohammad Ruhul and Mohammed, Nabeel and Rahman, Shafin. 2024. BanglaDialecto: An End-to-End AI-Powered Regional Speech Standardization. IEEE. PID https://doi.org/10.1109/BigData62323.2024.10826131. [↗](https://doi.org/10.1109/bigdata62323.2024.10826131)
30. Sultana, Nusrat and Yasmin, Rumana and Mallik, Bijon and Uddin, Mohammad Shorif. 2025. ONUBAD: A comprehensive dataset for automated conversion of Bangla regional dialects into standard Bengali dialect. Elsevier. PID https://doi.org/10.17632/6ft99kf89b.2.12.1. Dataset Sample Distribution Dataset Source Proprietary Dialect Dataset Vashantor BIDWESH Kothon BanglaDialecto Bangla Regonal Text Corpus Bangla Dialect Dataset Feniverse ChatgaiyyaAlap Onubad Sylheti Dialects into Standard SylhetiCAP Total Total Count 17, 004 9892 8985 7989 4477 4462 4038 4035 4007 2565 1188 24 68, 666 Covered Regions Chittagong, Comilla, Habiganj, Kishoreganj, Narsingdi, Noakhali, Rangpur, Tangail Barishal, Chittagong, Mymensingh, Noakhali, Sylhet Barishal, Chittagong, Noakhali Chittagong Noakhali Barishal, Khulna, Narail, Rangpur Barishal, Chittagong, Mymensingh, Rajshahi, Rangpur, Sylhet Feni Chittagong Barishal, Chittagong, Sylhet Sylhet Sylhet 12 Table 3: Dataset-wise total sample count and covered districts (sorted in descending order). Regions Chittagong Feni Kishoreganj Sylhet Mymensing Narsingdi Barishal Khulna Rangpur Habiganj Tangail Cumilla Rajshashi Noakhali Narail [↗](https://doi.org/10.17632/6ft99kf89b.2.12.1)