Aryaman Arora

Hi, I’m Aryaman! I’m an undergrad at Georgetown University studying computer science and linguistics. I do computational linguistics research as a member of NERT, Nathan Schneider’s research group at Georgetown. For this summer, I am a research intern at ETH Zürich with Ryan Cotterell. I’m excited about making technologies for South Asian languages and doing linguistic work on Indo-Aryan languages.

I love learning languages—besides being a native speaker of Hindi and English, I’m comfortable with Urdu and Punjabi and am actively learning Mandarin and Sanskrit. I’m an admin on the English Wiktionary where I manage South Asian language documentation. I also run a South Asian linguistics reading group, Moḻi-maṇḍala.

I like to bicycle (though not very well), read (in English and Hindi, but the latter not very well), and eat unreasonable amounts of food relative to my size. I’d also say I like to go places, but we’re in the middle of a pandemic. With my friend Ananya Gulati, I founded Washingtutors, an online tutoring non-profit for students in the D.C. Public School system (in which I spent my best years of high school, at the School Without Walls).

Check out my CV.

news

Jul 1, 2021	Started my internship at ETH Zürich with Ryan Cotterell!
Feb 15, 2021	I’m presenting “SNACS Annotation of Case Markers and Adpositions in Hindi” at SCiL 2021.
Nov 16, 2020	Attending EMNLP and presenting at SIGTYP.
Aug 27, 2020	My research is in the news! “First-Year Student Presented Paper at Prestigious Computational Linguistics Conference”
Aug 3, 2020	Starting undergrad at Georgetown.

selected publications

Bhāṣācitra: Visualising the dialect geography of South Asia

Aryaman Arora, Adam Farris, Gopalakrishnan R, Samopriya Basu

In Proceedings of the 2nd International Workshop on Computational Approaches to Historical Language Change, 2021

Abs arXiv

We present Bhāṣācitra, a dialect mapping system for South Asia built on a database of linguistic studies of languages of the region annotated for topic and location data. We analyse language coverage and look towards applications to typology by visualising example datasets. The application is not only meant to be useful for feature mapping, but also serves as a new kind of interactive bibliography for linguists of South Asian languages.
SNACS Annotation of Case Markers and Adpositions in Hindi

Aryaman Arora, Nitin Venkateswaran, Nathan Schneider

In Proceedings of the Society for Computation in Linguistics, 2021

Abs HTML Code

We present in-progress annotation of semantic relations expressed through adpositions and case markers in a Hindi corpus. We used the multilingual SNACS annotation scheme, which has been applied to a variety of typologically diverse languages. Annotation problems in Hindi are examined and used to suggest changes to SNACS. We look towards finalizing the corpus and using it for future work in typology and semantic role-dependent tasks.
Supervised Grapheme-to-Phoneme Conversion of Orthographic Schwas in Hindi and Punjabi

Aryaman Arora, Luke Gessler, Nathan Schneider

In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, 2020

Abs arXiv ACL Code Slides

Hindi grapheme-to-phoneme (G2P) conversion is mostly trivial, with one exception: whether a schwa represented in the orthography is pronounced or unpronounced (deleted). Previous work has attempted to predict schwa deletion in a rule-based fashion using prosodic or phonetic analysis. We present the first statistical schwa deletion classifier for Hindi, which relies solely on the orthography as the input and outperforms previous approaches. We trained our model on a newly-compiled pronunciation lexicon extracted from various online dictionaries. Our best Hindi model achieves state of the art performance, and also achieves good performance on a closely related language, Punjabi, without modification.