Hi! I'm a final-year PhD Candidate in the Department of Linguistics at Stanford University, advised by Dan Jurafsky.
My research focuses on data-centric approaches to creating speech and language technologies for digitally-underserved languages and user populations.
In recent work, I have been examining how we can improve the adaptability and efficiency of large, pre-trained speech models for use in low-resource/-compute scenarios, such as those in endangered language documentation projects.
In summer 2023, I worked as a Machine Learning Engineer Intern at rime, developing human-in-the-loop training data acquisition pipelines for building multi-speaker conversational text-to-speech (TTS) models capable of generating a range of diverse accents.
In the summer prior, I interned with the Siri and Language Technologies (SaLT) team at Apple to investigate data selection approaches for improving rare-word speech recognition.
I am also an Affiliate Member of the ARC Centre of Excellence for the Dynamics of Language (CoEDL) and a Visiting Fellow at the Australian National University (ANU), where I was based prior to Stanford and, where I developed tools and workflows to help clean, process, and transform lexicographical data from several Australian Aboriginal languages (Arabana, Kaytetye, Warlpiri, Warumungu).
To pay forward the benefits I received from tutoring as a young immigrant kid (lived in 8 countries by 21), I volunteer as a maths tutor in the East Bay with Refugee & Immigrant Transitions. If you are from a historically underrepresented group and are interested in speech and language technology research, feel very free to e-mail if you'd like a chat!
B.A. (Hons. I), French
B.Sc., Mathematics/B.A., Philosophy; French
Other academic experience
Summer School. ARC Centre of Excellence for the Dynamics of Language, Melbourne, Australia
Summer School. ARC Centre of Excellence for the Dynamics of Language, Sydney, Australia
Machine Learning Summer School. Sydney, Australia
Winter School on Multilingualism across the Lifespan. Fribourg, Switzerland
Exchange semester. University of Lausanne, Switzerland
Visiting Fellow. College of Asia & the Pacific. Australian National University. Canberra, Australia
Affiliate Member. ARC Centre of Excellence for the Dynamics of Language
Visitor. College of Arts & Social Sciences. Australian National University. Canberra, Australia
Field, A., Verma, P., San, N.
, Eberhardt, J., & Jurafsky, D. (2023). Developing Speech Processing Pipelines for Police Accountability. In INTERSPEECH 2023
. International Speech Communication Association. Available on https://arxiv.org/pdf/2306.06086.pdf
Bartelds, M., San, N.
, McDonnell, B., Jurafsky, D., & Wieling, M. (2023). Making More of Little Data: Improving Low-Resource Automatic Speech Recognition Using Data Augmentation. In Proceedings of the Association for Computational Linguistics conference 2023 (ACL2023)
. Association for Computational Linguistics. Available on https://arxiv.org/pdf/2305.10951.pdf
, Bartelds, M., Billings, B., De Falco, E., H., F., Safri, J., Sahrozi, W., Foley, B., McDonnell, B., & Jurafsky, D. (2023). Leveraging supplementary text data to kick-start automatic speech recognition system development with limited transcriptions. In Proceedings of the Sixth Workshop on the Use of Computational Methods in the Study of Endangered Languages (ComputEL-6)
. Association for Computational Linguistics. Available on https://aclanthology.org/2023.computel-1.1/
, Bartelds, M., Ògúnrèmí, T., Mount, A., Thompson, R., Higgins, M., Barker, R., Simpson, J., & Jurafsky, D. (2022). Automated speech tools for helping communities process restricted-access corpora for language revival efforts. In Proceedings of the Fifth Workshop on the Use of Computational Methods in the Study of Endangered Languages (ComputEL-5)
(pp. 41–51). Association for Computational Linguistics. Available on https://aclanthology.org/2022.computel-1.6
, *Bartelds, M., Browne, M., Clifford, L., Gibson, F., Mansfield, J., Nash, D., Simpson, J., Turpin, M., Vollmer, M., Wilmoth, S., & Jurafsky, D. (2021). Leveraging pre-trained representations to improve access to untranscribed speech from endangered languages. In Proceedings of the 2021 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU)
. doi: 10.1109/ASRU51503.2021.9688301
, & Turpin, M. (2021). Text-setting in Kaytetye. In R. Bennett, R. Bibbs, M. Loren Brinkerhoff, M. J. Kaplan, S. Rich, A. Rysling, N. van Handel, & M. Wax Cavallaro (Eds.), Supplemental Proceedings of the 2020 Annual Meeting on Phonology
(pp. 1–9). doi: 10.3765/amp.v9i0.4911
van Esch, D., Foley, B., & San, N.
(2019). Future Directions in Technological Support for Language Documentation. In M. Silfverberg (Ed.), Proceedings of the 3rd Workshop on Computational Methods for Endangered Languages (ComputEL-3)
(Vol. 1). Available on https://scholar.colorado.edu/scil-cmel/vol1/iss1/3
Foley, B., Arnold, J., Coto-Solano, R., Durantin, G., Ellison, T. M., van Esch, D., Heath, S., Kratochvíl, F., Maxwell-Smith, Z., Nash, D., Olsson, O., Richards, M., San, N.
, Stoakes, H., Thieberger, N., & Wiles, J. (2018). Building Speech Recognition Systems for Language Documentation: The CoEDL Endangered Language Pipeline and Inference System (Elpis). In S. S. Agrawal (Ed.), The 6th Intl. Workshop on Spoken Language Technologies for Under-Resourced Languages (SLTU)
(pp. 200–204). Available on https://www.isca-speech.org/archive/SLTU_2018/pdfs/Ben.pdf
Harvey, M., San, N.
, Proctor, M., Panther, F., & Turpin, M. (2023). The Kaytetye segmental inventory. Australian Journal of Linguistics
. doi: 10.1080/07268602.2023.2218270
Foley, B., van Esch, D., & San, N.
(2022). Managing transcription data for automatic speech recognition with Elpis. In A. Berez-Kroeker, B. McDonnell, E. Koller, & L. Collister (Eds.), The Open Handbook of Linguistic Data Management
. Cambridge, MA: MIT Press. doi: 10.7551/mitpress/12200.003.0041
Harvey, M., San, N.
, Carew, M., Strangways, S., Simpson, J., & Stockigt, C. (2019). Pre-stopping in Arabana. Australian Journal of Linguistics
(4). doi: 10.1080/07268602.2019.1643290
San, N. (2023). Improving access to language documentation corpora using self-supervised models for speech. Phonetics, Phonology and Psycholinguistics Forum (Phorum), UC Berkeley (February, 24) and the Department of Computational Linguistics, University of Zurich (April, 18).
San, N., & van Esch, D. (2019). Accelerating Transcription of Fieldwork Data using Machine Learning. Fieldwork Forum (FForum), UC Berkeley. October, 3.
San, N. (2018). Quantitative methods in linguistics: an overview. Department of Mathematics & Statistics, La Trobe University, Melbourne. February, 8.
Turpin, M., & San, N. (2022). The prosodic word in literary traditions of variable line length. Paper presented at the conference of the Annual Meeting of the Linguistics Society of America.
, Disbray, S., Foley, B., & Simpson, J. (2019). Towards an extensible, open-source picture dictionary template and processing system. Paper presented at the conference of the Australasian Association for Lexicography (AustraLex). Australian National University, Canberra, Australia. doi: 10.5281/zenodo.3404266
, Carne, M., Carew, M., Harvey, M., Hercus, L., & Simpson, J. (2018). An acoustic analysis of pre-stopping in Arabana. Paper presented at the 18th Australian Languages Workshop (ALW). Marysville, Victoria. doi: 10.5281/zenodo.3404271
San, N. (2017). A corpus-based approach to vocalic contrasts in Kaytetye. Paper presented at the Corpus Workshop of the Centre of Excellence for the Dynamics of Language. Melbourne, Australia.
San, N. (2015). Visualising the articulatory characteristics of Kaytetye coronal consonants. Paper presented at the Building Corpora for Australian Languages workshop. Australian National University, Canberra.
San, N., Proctor, M., Turpin, M., Harvey, M., Ringbauer, K., Ross, A., & Demuth, K. (2015). Variability of vowels in Kaytetye words. Paper presented at the Arandic Phonetics & Phonology Workshop. Alice Springs, NT, Australia.
San, N., Proctor, M., Turpin, M., Harvey, M., Ringbauer, K., Ross, A., & Demuth, K. (2015). An acoustic analysis of Kaytetye vowel variability. Paper presented at the 46th Annual Conference of the Australian Linguistic Society, Western Sydney University, Australia.
San, N., & Turpin, M. (2014). Acoustic correlates of stress in Kaytetye words. Paper presented at the Workshop on Word Stress & Accent. Leiden University, the Netherlands.
San, N., & Turpin, M. (2014). Acoustic correlates of stress in Kaytetye words. Paper presented at the 45th Annual Conference of the Australian Linguistic Society (ALS). University of Newcastle, Australia.
San, N., Bartelds, M., & Jurafsky, D. (2021). Improving Access to Untranscribed Speech by Leveraging Spoken Term Detection and Self-supervised Learning of Speech Representations. Invited non-archival extended abstract presented at SigTyp 2021.
San, N., & Turpin, M. (2020). Text-setting in Kaytetye. Poster presented at the 2020 Annual Meeting on Phonology (Online).
(2016). Using version control to facilitate a reproducible and collaborative workflow in acoustic phonetics. In C. Carignan & M. D. Tyler (Eds.), Proceedings of the 16th Australasian Speech Science and Technology Association conference
(pp. 341–344). Available on http://www.assta.org/sst/2016/papers/San_SST2016.pdf
San, N. (2013). The perception of French vowels by Australian-English-speaking learners. Poster presented at the 21st Annual Conference of the Australian Society of French Studies (ASFS). The University of Queensland, Brisbane, Australia.
Honours and Awards
Summer Research Scholarship (Linguistics). Australian National University
Research Training Pathway Scholarship, Year 2. Macquarie University
Research Training Pathway Scholarship, Year 1. Macquarie University
G.M. Grassie Memorial Prize. University of Queensland
Dean's Commendations for Academic Excellence. University of Queensland
Summer Research Scholarship (Linguistics). University of Queensland
Exchange student allowance. University of Lausanne, Switzerland
Travel grant for exchange studies. University of Queensland
Summer Research Scholarship (Mathematics). University of Queensland
Summer Research Scholarship (Mathematics). University of Queensland
San, N., Foley, B., Disbray, S., Turpin, M. and Simpson, J. Towards an extensible, open-source picture dictionary template and processing system. Transdisciplinary and Innovation Grant from the ARC Centre of Excellence for the Dynamics of Language. AUD 19,809
Harvey, M., Hercus, L., Carew, M. and San, N. Metrical prominence and pre-stopping in Arabana. Language Documentation Grant from the ARC Centre of Excellence for the Dynamics of Language. AUD 10,992
Warlpiri Lexicography Project. Australian National University. Supervisor: Jane Simpson
Developing Yerrampe: A Kaytetye-to-English multimedia dictionary. Sydney Conservatorium of Music. Supervisor: Myfany Turpin
Investigating tongue shaping of Kaytetye obstruents using ultrasound imaging. Macquarie University. Supervisors: Michael Proctor & Katherine Demuth
Examining the acoustic correlates of stress in Kaytetye words. University of Queensland. Supervisor: Myfany Turpin
Teaching, as Teaching Assistant
Phonetics, Winter Quarter (Instructor: Meghan Sumner)
Introduction to Phonology, Winter Quarter (Instructor: Arto Anttila)
Speech Physiology, Semester 1 (Instructor: Michael Proctor)
University of Queensland
Calculus & Linear Algebra I (4 Semesters; Instructor: Phil Isaac)
Calculus & Linear Algebra II (4 Semesters; Instructor: Phil Isaac)
Multivariate Calculus & Differential Equations, Semester 1 (Instructor: Phil Isaac)
Discrete Mathematics I, Semester 2 (Instructor: Murray Elder)
Teaching, as Workshop Instructor
Introduction to Elpis for developing Automatic Speech Recognition for Field Data. Co-presented with Daan van Esch at the Department of Linguistic and Cultural Evolution's workshop on Linguistic Analysis/Data Management at the Max Planck Institute for Evolutionary Anthropology in Leipzig, Germany. April, 4-5
Semi-automated transcription for Language Documentation with Elpis. Co-presented with Ben Foley, Nicholas Lambourne, and Daan van Esch at the 7th International Conference on Language Documentation and Conservation (Online). University of Hawai'i at Mānoa. March, 5-6
Transcription Acceleration for Language Documentation with ELPIS. Co-presented with Ben Foley and Daan van Esch at the 6th International Conference on Language Documentation and Conservation. University of Hawai'i at Mānoa. March, 1-2
Docker: An overview for linguists. Presented at the Transcription Acceleration Project workshop. Melbourne, Australia. February, 4
Visualise your own vowels: A short introduction to Praat for beginners. 1-day tutorial at the Summer School of the ARC Centre of Excellence for the Dynamics of Language. Canberra, Australia. November, 29
Formalising data structures with the Nearley toolkit. University of Queensland. September, 28
Introduction to data wrangling using R. University of Melbourne, Australia. May, 1
A beginner’s guide to programming in R and Matlab. Child Language Lab, Macquarie University. May, 4
LaTeX for Linguistics. School of Languages & Comparative Cultural Studies. University of Queensland. Weekly tutorials, April–May
Guest Lectures and Public Outreach
The building blocks of spoken language. Presentation to visiting final-year high schoolers at the National Youth Science Forum. Australian National University. Canberra, Australia. January, 10
Other Teaching Experience
Freelance English language instructor. Passmore College, Marburg, Germany
Foreign language teaching assistant. Ministry of Education, Aurillac, France
University of Queensland
Honours students representative, Semester 2. School of Languages and Cultures
Reviewer, Journal of the Canadian Acoustical Association
Volunteer tutor. Refugee and Immigrant Transitions. Oakland
Volunteer tutor. Berkeley Public Schools Fund
Volunteer tutor. Migrant and Refugee Settlement Services. Canberra
Volunteer. Refugee Action Coalition. Sydney
English (native), French (proficient, CEFR C1), German (conversational, CEFR B2), Burmese (heritage)
Familiarity with processing to/from SQL (MySQL/Postgres/MariaDB), XML, JSON, columnar tab-/comma-separated values, Toolbox lexicon files, Praat TextGrid annotation files; writing custom parsers for domain-specific data formats; developing and deploying data-cleaning and -validation pipelines