---
title: "Insights from Romanized Manipuri Social Media — Salice 2026"
description: "Manipuri (Meitei) is a Tibeto-Burman language of Northeast India that is traditionally written in Manipuri-Bengali or Meitei Mayek script."
authors: ["Maisang Kamei Salice", "Sanasam Ranbir Singh", "Priyankoo Sarmah"]
year: 2026
venue: "Proceedings of the Fifteenth Language Resources and Evaluation Conference"
source: "https://aclanthology.org/2026.lrec-1.147/"
doi: "10.63317/3uqwwrf7jvvo"
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/en/papers/salice-2026-insights-romanized-manipuri-social-media/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/salice-2026-insights-romanized-manipuri-social-media/)

[home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2026](https://md.research.elamur.ai/free/en/papers/2026/page/5/) / Insights from Romanized Manipuri Social Media Text: A Transliteration Corpus and Variation Analysis

# Insights from Romanized Manipuri Social Media Text: A Transliteration Corpus and Variation Analysis

**Authors:** Maisang Kamei Salice, [Sanasam Ranbir Singh](https://md.research.elamur.ai/free/en/authors/sanasam-ranbir-singh/), [Priyankoo Sarmah](https://md.research.elamur.ai/free/en/authors/priyankoo-sarmah/) · **Year:** 2026 · **Venue:** Proceedings of the Fifteenth Language Resources and Evaluation Conference · **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.lrec-1.147/) · [PDF](https://aclanthology.org/2026.lrec-1.147.pdf) · [DOI 10.63317/3uqwwrf7jvvo](https://doi.org/10.63317/3uqwwrf7jvvo) · [OpenAlex](https://openalex.org/W7160218853)

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

## Summary

Manipuri (Meitei) is a Tibeto-Burman language of Northeast India that is traditionally written in Manipuri-Bengali or Meitei Mayek script. On social media, however, speakers usually type it in Latin letters without any agreed spelling convention. Existing transliteration resources such as Aksharantar contain only clean, canonical pairs, so they miss the noise found in real user text.

The authors built a word-level parallel corpus that maps Romanized Manipuri from YouTube comments and live chats to Manipuri-Bengali script, annotated by native speakers. They examine variation on two levels:
- **Character mappings**, which are inconsistent (e.g., /p/ typed as p, b or f).
- **Stylistic habits**, such as dropped vowels, digit substitutions and stretched letters.

They also compare user spellings with formal schemes such as ISO 15919, ITRANS and textbook romanization. Transformer models with character-level and BPE subword inputs are then evaluated, followed by a detailed error analysis. Character-level modeling performs better, but vowels, digits and spacing remain hard. The resource fills a gap for normalizing noisy text in an underserved language, and the observed patterns likely apply to other romanized low-resource languages.

## Key points

- **Data collection:** Comments were gathered via the YouTube Data API using 66 search queries (39 channel names, 27 general keywords). News & Public Affairs makes up 24.2% of the content, Culture & Knowledge accounts for 31.8%, and Entertainment holds the largest share at 43.9%. Collection took place July–September 2022 and covers content from 2019–2022.
- **Annotation:** 27 native Manipuri speakers passed a qualification set of at least 20 sentences and worked on a custom online platform with two-stage quality control. The dataset is available on request.
- **Corpus statistics:** The corpus follows Zipf's and Heaps' laws. Vowels show wider and less skewed spelling variation than consonants, and vowel diphthongs depart most from standardized romanization.
- **Model results:** Character-level Transformer models outperform BPE subword models. Most errors stem from vowel grapheme mismatches.
- **Error analysis:** Errors are grouped into categories including gemination, contextual ambiguity, space insertion or omission, missing characters, onset–coda ambiguity and digits. The model handles the repetition marker "2" well but struggles with other digits.

*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: [Linguistics & languages](https://md.research.elamur.ai/free/en/shelves/linguistics/)

## Cite

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

```bibtex
@inproceedings{salice2026insights,
  title = {{Insights from Romanized Manipuri Social Media Text: A Transliteration Corpus and Variation Analysis}},
  author = {Salice, Maisang Kamei and Singh, Sanasam Ranbir and Sarmah, Priyankoo},
  booktitle = {Proceedings of the Fifteenth Language Resources and Evaluation Conference},
  year = {2026},
  doi = {10.63317/3uqwwrf7jvvo},
  url = {https://doi.org/10.63317/3uqwwrf7jvvo}
}
```

- [← previous Information Asymmetry across Language Varieties: A Case Study on Cantonese-Mandarin and Bavarian-German QA](https://md.research.elamur.ai/free/en/papers/pei-2026-information-asymmetry-across-language-varieties/)
- [next → Insights from Transfer Learning Experiments with Word-in-Context and Word Sense Disambiguation Models](https://md.research.elamur.ai/free/en/papers/mujko-2026-insights-transfer-learning-experiments-word/)
- [↑ back to papers of 2026](https://md.research.elamur.ai/free/en/papers/2026/page/5/)

## References

32 references, 4 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Umair Z Ahmed, Kalika Bali, Monojit Choudhury, and Sowmya VB. 2011. Challenges in designing input method editors for Indian languages: The role of word-origin and context. In Proceedings of the Workshop on Advances in Text Input Methods (WTIM 2011), pages 1–9, Chiang Mai, Thailand. Asian Federation… [find](https://scholar.google.com/scholar?q=Umair+Z+Ahmed%2C+Kalika+Bali%2C+Monojit+Choudhury%2C+and+Sowmya+VB%2E+2011%2E+Challenges+in+designing+input+method+editors+for+Indian+languages%3A+The+role+of+word%2Dorigin+and+context%2E+In+Proceedings+of+the+Worksh)
2. Hemanta Baruah, Sanasam Ranbir Singh, and Priyankoo Sarmah. 2024a. Assamesebacktranslit: Back transliteration of romanized assamese social media text. In Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024), pages 1… [find](https://scholar.google.com/scholar?q=Hemanta+Baruah%2C+Sanasam+Ranbir+Singh%2C+and+Priyankoo+Sarmah%2E+2024a%2E+Assamesebacktranslit%3A+Back+transliteration+of+romanized+assamese+social+media+text%2E+In+Proceedings+of+the+2024+Joint+International+Co)
3. Hemanta Baruah, Sanasam Ranbir Singh, and Priyankoo Sarmah. 2024b. Transliteration characteristics in romanized assamese language social media text and machine transliteration. ACM Transactions on Asian and Low-Resource Language Information Processing, 23(2):1–36. [find](https://scholar.google.com/scholar?q=Hemanta+Baruah%2C+Sanasam+Ranbir+Singh%2C+and+Priyankoo+Sarmah%2E+2024b%2E+Transliteration+characteristics+in+romanized+assamese+language+social+media+text+and+machine+transliteration%2E+ACM+Transactions+on+Asi)
4. Irshad Ahmad Bhat, Vandan Mujadia, Aniruddha Tammewar, Riyaz Ahmad Bhat, and Manish Shrivastava. 2014. Iiit-h system submission for fire2014 shared task on transliterated search. In Proceedings of the 6th Annual Meeting of the Forum for Information Retrieval Evaluation, FIRE ’14, page 48–53, New Yor… [find](https://scholar.google.com/scholar?q=Irshad+Ahmad+Bhat%2C+Vandan+Mujadia%2C+Aniruddha+Tammewar%2C+Riyaz+Ahmad+Bhat%2C+and+Manish+Shrivastava%2E+2014%2E+Iiit%2Dh+system+submission+for+fire2014+shared+task+on+transliterated+search%2E+In+Proceedings+of+the)
5. Ann Bies, Zhiyi Song, Mohamed Maamouri, Stephen Grimes, Haejoong Lee, Jonathan Wright, Stephanie Strassel, Nizar Habash, Ramy Eskander, and Owen Rambow. 2014. Transliteration of arabizi into arabic orthography: Developing a parallel annotated arabizi-arabic script sms/chat corpus. In Proceedings of… [find](https://scholar.google.com/scholar?q=Ann+Bies%2C+Zhiyi+Song%2C+Mohamed+Maamouri%2C+Stephen+Grimes%2C+Haejoong+Lee%2C+Jonathan+Wright%2C+Stephanie+Strassel%2C+Nizar+Habash%2C+Ramy+Eskander%2C+and+Owen+Rambow%2E+2014%2E+Transliteration+of+arabizi+into+arabic+or)
6. Kunal Chakma and Amitava Das. 2014. Revisiting automatic transliteration problem for code-mixed romanized indian social media text. Social-India, 2014:42. [find](https://scholar.google.com/scholar?q=Kunal+Chakma+and+Amitava+Das%2E+2014%2E+Revisiting+automatic+transliteration+problem+for+code%2Dmixed+romanized+indian+social+media+text%2E+Social%2DIndia%2C+2014%3A42%2E)
7. Md Fahim, Fariha Shifat, Fabiha Haider, Deeparghya Barua, Md Sourove, Md Ishmam, and Md Bhuiyan. 2024. BanglaTLit: A benchmark dataset for back-transliteration of Romanized Bangla. In Findings of the Association for Computational Linguistics: EMNLP 2024, pages 14656–14672. [find](https://scholar.google.com/scholar?q=Md+Fahim%2C+Fariha+Shifat%2C+Fabiha+Haider%2C+Deeparghya+Barua%2C+Md+Sourove%2C+Md+Ishmam%2C+and+Md+Bhuiyan%2E+2024%2E+BanglaTLit%3A+A+benchmark+dataset+for+back%2Dtransliteration+of+Romanized+Bangla%2E+In+Findings+of+the+)
8. Philip Gage. 1994. A new algorithm for data compression. C Users Journal, 12(2):23–38. [find](https://scholar.google.com/scholar?q=Philip+Gage%2E+1994%2E+A+new+algorithm+for+data+compression%2E+C+Users+Journal%2C+12%282%29%3A23%E2%80%9338%2E)
9. Nizar Habash, Mona Diab, and Owen Rambow. 2012. Conventional orthography for dialectal Arabic. In Proceedings of the Eighth International Conference on Language Resources and Evaluation (LREC’12), pages 711–718, Istanbul, Turkey. European Language Resources Association (ELRA). [find](https://scholar.google.com/scholar?q=Nizar+Habash%2C+Mona+Diab%2C+and+Owen+Rambow%2E+2012%2E+Conventional+orthography+for+dialectal+Arabic%2E+In+Proceedings+of+the+Eighth+International+Conference+on+Language+Resources+and+Evaluation+%28LREC%E2%80%9912%29%2C+pag) [our card](https://md.research.elamur.ai/free/en/papers/habash-2012-conventional-orthography-dialectal-arabic/)
10. Harold Stanley Heaps. 1978. Information Retrieval: Computational and Theoretical Aspects. Academic Press, New York. [find](https://scholar.google.com/scholar?q=Harold+Stanley+Heaps%2E+1978%2E+Information+Retrieval%3A+Computational+and+Theoretical+Aspects%2E+Academic+Press%2C+New+York%2E)
11. Dan Hendrycks and Kevin Gimpel. 2023. Gaussian error linear units (gelus). [find](https://scholar.google.com/scholar?q=Dan+Hendrycks+and+Kevin+Gimpel%2E+2023%2E+Gaussian+error+linear+units+%28gelus%29%2E)
12. Daniel Jurafsky and James H. Martin. 2025. Speech and Language Processing: An Introduction to Natural Language Processing, Computational Linguistics, and Speech Recognition with Language Models, 3rd edition. Online manuscript released August 24, 2025. [find](https://scholar.google.com/scholar?q=Daniel+Jurafsky+and+James+H%2E+Martin%2E+2025%2E+Speech+and+Language+Processing%3A+An+Introduction+to+Natural+Language+Processing%2C+Computational+Linguistics%2C+and+Speech+Recognition+with+Language+Models%2C+3rd+e)
13. Sarvnaz Karimi, Falk Scholer, and Andrew Turpin. 2011. Machine transliteration survey. ACM Computing Surveys (CSUR), 43(3):1–46. [find](https://scholar.google.com/scholar?q=Sarvnaz+Karimi%2C+Falk+Scholer%2C+and+Andrew+Turpin%2E+2011%2E+Machine+transliteration+survey%2E+ACM+Computing+Surveys+%28CSUR%29%2C+43%283%29%3A1%E2%80%9346%2E)
14. Kamaljeet Kaur and Parminder Singh. 2014. Review of machine transliteration techniques. International Journal of Computer Applications, 107(20):13–16. [find](https://scholar.google.com/scholar?q=Kamaljeet+Kaur+and+Parminder+Singh%2E+2014%2E+Review+of+machine+transliteration+techniques%2E+International+Journal+of+Computer+Applications%2C+107%2820%29%3A13%E2%80%9316%2E)
15. Lenin Laitonjam, Loitongbam Gyanendro Singh, and Sanasam Ranbir Singh. 2018. Transliteration of english loanwords and named-entities to manipuri: Phoneme vs grapheme representation. In 2018 International Conference on Asian Language Processing (IALP), pages 255–260. IEEE. [find](https://scholar.google.com/scholar?q=Lenin+Laitonjam%2C+Loitongbam+Gyanendro+Singh%2C+and+Sanasam+Ranbir+Singh%2E+2018%2E+Transliteration+of+english+loanwords+and+named%2Dentities+to+manipuri%3A+Phoneme+vs+grapheme+representation%2E+In+2018+Internatio)
16. Lenin Laitonjam and Sanasam Ranbir Singh. 2022. A hybrid machine transliteration model based on multi-source encoder–decoder framework: English to manipuri. SN Computer Science, 3(2):125. [find](https://scholar.google.com/scholar?q=Lenin+Laitonjam+and+Sanasam+Ranbir+Singh%2E+2022%2E+A+hybrid+machine+transliteration+model+based+on+multi%2Dsource+encoder%E2%80%93decoder+framework%3A+English+to+manipuri%2E+SN+Computer+Science%2C+3%282%29%3A125%2E)
17. WMP Liwera and L Ranathunga. 2020. Combination of trigram and rule-based model for singlish to sinhala transliteration by focusing social media text. In 2020 From Innovation to Impact (FITI), volume 1, pages 1–5. IEEE. [find](https://scholar.google.com/scholar?q=WMP+Liwera+and+L+Ranathunga%2E+2020%2E+Combination+of+trigram+and+rule%2Dbased+model+for+singlish+to+sinhala+transliteration+by+focusing+social+media+text%2E+In+2020+From+Innovation+to+Impact+%28FITI%29%2C+volume+1)
18. Yash Madhani, Sushane Parthan, Priyanka Bedekar, Gokul Nc, Ruchi Khapra, Anoop Kunchukuttan, Pratyush Kumar, and Mitesh M Khapra. 2022. Aksharantar: Open indic-language transliteration datasets and models for the next billion users. arXiv preprint arXiv:2205.03018, pages 40–57. [find](https://scholar.google.com/scholar?q=Yash+Madhani%2C+Sushane+Parthan%2C+Priyanka+Bedekar%2C+Gokul+Nc%2C+Ruchi+Khapra%2C+Anoop+Kunchukuttan%2C+Pratyush+Kumar%2C+and+Mitesh+M+Khapra%2E+2022%2E+Aksharantar%3A+Open+indic%2Dlanguage+transliteration+datasets+and+mo)
19. Sabina Mammadzada. 2023. A review of existing transliteration approaches and methods. International Journal of Multilingualism, 20(3):1052–1066. [find](https://scholar.google.com/scholar?q=Sabina+Mammadzada%2E+2023%2E+A+review+of+existing+transliteration+approaches+and+methods%2E+International+Journal+of+Multilingualism%2C+20%283%29%3A1052%E2%80%931066%2E)
20. Jayashree Nair and Riyaz Ahammed. 2021. English to indian language and back transliteration with phonetic transcription for computational linguistics tools based on conventional transliteration schemes. In 2021 Fourth International Conference on Electrical, Computer and Communication Technologies (I… [find](https://scholar.google.com/scholar?q=Jayashree+Nair+and+Riyaz+Ahammed%2E+2021%2E+English+to+indian+language+and+back+transliteration+with+phonetic+transcription+for+computational+linguistics+tools+based+on+conventional+transliteration+scheme)
21. Kishorjit Nongmeikapam and Sivaji Bandyopadhyay. 2012. A transliteration of crf based manipuri pos tagging. Procedia Technology, 6:582–589. [find](https://scholar.google.com/scholar?q=Kishorjit+Nongmeikapam+and+Sivaji+Bandyopadhyay%2E+2012%2E+A+transliteration+of+crf+based+manipuri+pos+tagging%2E+Procedia+Technology%2C+6%3A582%E2%80%93589%2E)
22. Kishorjit Nongmeikapam, Ningombam Herojit Singh, Sonia Thoudam, and Sivaji Bandyopadhyay. 2011. Manipuri transliteration from bengali script to meitei mayek: A rule based approach. In International Conference on Information Systems for Indian Languages, pages 195–198. Springer. [find](https://scholar.google.com/scholar?q=Kishorjit+Nongmeikapam%2C+Ningombam+Herojit+Singh%2C+Sonia+Thoudam%2C+and+Sivaji+Bandyopadhyay%2E+2011%2E+Manipuri+transliteration+from+bengali+script+to+meitei+mayek%3A+A+rule+based+approach%2E+In+International+Co)
23. Santanu Pal, Partha Pakray, Sahinur Rahman Laskar, Lenin Laitonjam, Vanlalmuansangi Khenglawt, Sunita Warjri, Pankaj Kundan Dadure, and Sandeep Kumar Dash. 2023. Findings of the wmt 2023 shared task on low-resource indic language translation. In Proceedings of the Eighth Conference on Machine Transl… [find](https://scholar.google.com/scholar?q=Santanu+Pal%2C+Partha+Pakray%2C+Sahinur+Rahman+Laskar%2C+Lenin+Laitonjam%2C+Vanlalmuansangi+Khenglawt%2C+Sunita+Warjri%2C+Pankaj+Kundan+Dadure%2C+and+Sandeep+Kumar+Dash%2E+2023%2E+Findings+of+the+wmt+2023+shared+task+o) [our card](https://md.research.elamur.ai/free/en/papers/pal-2023-findings-wmt-2023-shared-task/)
24. 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. [find](https://scholar.google.com/scholar?q=Kishore+Papineni%2C+Salim+Roukos%2C+Todd+Ward%2C+and+Wei%2DJing+Zhu%2E+2002%2E+Bleu%3A+a+method+for+automatic+evaluation+of+machine+translation%2E+In+Proceedings+of+the+40th+annual+meeting+of+the+Association+for+Comp) [our card](https://md.research.elamur.ai/free/en/papers/papineni-2002-bleu-method-automatic-evaluation-machine/)
25. Pranav Rajpurkar, Jian Zhang, Konstantin Lopyrev, and Percy Liang. 2016. Squad: 100,000+ questions for machine comprehension of text. arXiv preprint arXiv:1606.05250. [find](https://scholar.google.com/scholar?q=Pranav+Rajpurkar%2C+Jian+Zhang%2C+Konstantin+Lopyrev%2C+and+Percy+Liang%2E+2016%2E+Squad%3A+100%2C000%2B+questions+for+machine+comprehension+of+text%2E+arXiv+preprint+arXiv%3A1606%2E05250%2E) [our card](https://md.research.elamur.ai/free/en/papers/rajpurkar-2016-squad-100-000-questions-machine/)
26. Brian Roark, Lawrence Wolf-Sonkin, Christo Kirov, Sabrina J. Mielke, Cibu Johny, Isin Demirsahin, and Keith Hall. 2020. Processing South Asian languages written in the Latin script: the Dakshina dataset. In Proceedings of the Twelfth Language Resources and Evaluation Conference, pages 2413–2423, Mar… [find](https://scholar.google.com/scholar?q=Brian+Roark%2C+Lawrence+Wolf%2DSonkin%2C+Christo+Kirov%2C+Sabrina+J%2E+Mielke%2C+Cibu+Johny%2C+Isin+Demirsahin%2C+and+Keith+Hall%2E+2020%2E+Processing+South+Asian+languages+written+in+the+Latin+script%3A+the+Dakshina+datas)
27. Inder Singh. 1975. Manipuri phonetic reader. Central Institute of Indian Languages, Mysore. [find](https://scholar.google.com/scholar?q=Inder+Singh%2E+1975%2E+Manipuri+phonetic+reader%2E+Central+Institute+of+Indian+Languages%2C+Mysore%2E)
28. Leihaorambam Sarbajit Singh, Kabita Thaoroijam, and Pradip Kumar Das. 2007. Written manipuri (meiteiron)–phoneme to grapheme correspondence. Language in India Strength for Today and Bright Hope for Tomorrow Volume 7 : 6 June 2007. [find](https://scholar.google.com/scholar?q=Leihaorambam+Sarbajit+Singh%2C+Kabita+Thaoroijam%2C+and+Pradip+Kumar+Das%2E+2007%2E+Written+manipuri+%28meiteiron%29%E2%80%93phoneme+to+grapheme+correspondence%2E+Language+in+India+Strength+for+Today+and+Bright+Hope+for+To)
29. Thoudam Doren Singh. 2012. Bidirectional bengali script and meetei mayek transliteration of web based manipuri news corpus. In Proceedings of the 3rd Workshop on South and Southeast Asian Natural Language Processing, pages 181–190. [find](https://scholar.google.com/scholar?q=Thoudam+Doren+Singh%2E+2012%2E+Bidirectional+bengali+script+and+meetei+mayek+transliteration+of+web+based+manipuri+news+corpus%2E+In+Proceedings+of+the+3rd+Workshop+on+South+and+Southeast+Asian+Natural+Lang)
30. Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N Gomez, Łukasz Kaiser, and Illia Polosukhin. 2017. Attention is all you need. Advances in neural information processing systems, 30. [find](https://scholar.google.com/scholar?q=Ashish+Vaswani%2C+Noam+Shazeer%2C+Niki+Parmar%2C+Jakob+Uszkoreit%2C+Llion+Jones%2C+Aidan+N+Gomez%2C+%C5%81ukasz+Kaiser%2C+and+Illia+Polosukhin%2E+2017%2E+Attention+is+all+you+need%2E+Advances+in+neural+information+processing+)
31. MK Vathsala and Ganga Holi. 2020. Rnn based machine translation and transliteration for twitter data. International Journal of Speech Technology, 23(3):499–504. [find](https://scholar.google.com/scholar?q=MK+Vathsala+and+Ganga+Holi%2E+2020%2E+Rnn+based+machine+translation+and+transliteration+for+twitter+data%2E+International+Journal+of+Speech+Technology%2C+23%283%29%3A499%E2%80%93504%2E)
32. George Kingsley Zipf. 1949. Human Behavior and the Principle of Least Effort: An Introduction to Human Ecology. Addison-Wesley Press, Cambridge, MA. [find](https://scholar.google.com/scholar?q=George+Kingsley+Zipf%2E+1949%2E+Human+Behavior+and+the+Principle+of+Least+Effort%3A+An+Introduction+to+Human+Ecology%2E+Addison%2DWesley+Press%2C+Cambridge%2C+MA%2E)