Using artificial-intelligence tools to make LaTeX content accessible to blind readers
Open accessAbstract
PDF documents containing formulae generated by L A T E X are usually not accessible by assistive technologies for visually impaired people (i.e., by screen readers and braille displays).The L A T E X package axessibility.stythat we have developed alleviates this issue, allowing one to create PDF documents where the formulae are read by these assistive technologies, since it automatically generates hidden comments in the PDF document (using the /ActualText attribute) in correspondence to each formula.This actual text is hidden in the PDF document, but the screen readers JAWS, NVDA and VoiceOver read it correctly.Moreover, we have created NVDA and Jaws dictionaries (in English and in Italian) that provide reading in the natural language in case the user does not know L A T E X commands.The package does not generate PDF/UA.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
Kortemeyer (2023) found promising results in using GPT-4 to convert LaTeX documents containing equations into screen-reader-accessible HTML.
“For instance, Kortemeyer (2023) found promising results in using GPT-4 to convert LaTeX documents containing equations into screen-reader-accessible HTML.”
From Exploring the role of generative AI in higher education: Semi-structured interviews with students with disabilities (2025) · cites this paper as Kortemeyer (2023)
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Cites
No references could be resolved yet.
Research works printed but not linked: 18
- Y. Borodin, J. P. Bigham, G. Dausch, and I. Ramakrish- nan, in Proceedings of the 2010 International Cross Dis- ciplinary Conference on Web Accessibility (W4A) (2010), pp. 1-10.
- J. T. Nganji, Library & Information Science Research 37, 254 (2015).
- A. Petz and K. Miesenberger, in AAATE Conf. (2015), pp. 27-31.
- J. Gilligan, in Universal Access in Human-Computer In- teraction. Applications and Practice: 14th International Conference, UAHCI 2020, Held as Part of the 22nd HCI International Conference, HCII 2020, Copenhagen, Denmark, July 19-24, 2020, Proceedings, Part II 22 (Springer, 2020), pp. 184-199.
- D. Ahmetovic, T. Armano, C. Bernareggi, M. Berra, A. Capietto, S. Coriasco, N. Murru, A. Ruighi, and E. Taranto, in Proceedings of the 20th international ACM SIGACCESS conference on computers and accessibility (2018), pp. 352-354.
- A. Manzoor, S. Arooj, S. Zulfiqar, M. Parvez, S. Shahid, and A. Karim, in Proceedings of the 2019 CHI conference on human factors in computing systems (2019), pp. 1-12.
- A. Darvishy, in Computers Helping People with Spe- cial Needs: 16th International Conference, ICCHP 2018, Linz, Austria, July 11-13, 2018, Proceedings, Part I 16 (Springer, 2018), pp. 113-116.
- A. Jembu Rajkumar, J. Lazar, J. B. Jordan, A. Darvishy, and H.-P. Hutter, in Proceedings of the 53rd Hawaii International Conference on System Sciences (HICSS) (2020).
- D. Cervone, Notices of the AMS 59, 312 (2012).
- D. Cervone, P. Krautzberger, and V. Sorge, The Jour- nal on Technology and Persons with Disabilities 4, 167 (2016).
- V. Sorge, C. Chen, T. Raman, and D. Tseng, in Pro- ceedings of the 11th web for all conference (2014), pp. 1-10.
- G. Kortemeyer, Phys. Rev. Phys. Educ. Res. 19, 010132 (2023), URL https://link.aps.org/doi/10. 1103/PhysRevPhysEducRes.19.010132. · DOI
- R. Fateman, Journal of Symbolic Computation 25 (1998).
- A. Ogawa, TUGboat 22, 131 (2001).
- Z. Walczak, TUGboat 20, 176 (2008).
- J. Fenn, The PracTEX Journal 4, 2006 (2006).
- S. Iwarsson and A. Ståhl, Disability and rehabilitation 25, 57 (2003).
- H. Persson, H. Åhman, A. A. Yngling, and J. Gullik- sen, Universal Access in the Information Society 14, 505 (2015).
Cited by
Sorted by citations
Pierrès, O., Darvishy, A., & Christen, M. (2025) · Education and Information Technologies · doi:10.1007/s10639-024-13134-8 · Cited by 48 · see citing works
- Authors
- Kortemeyer, G.
- Affiliations
- ETH Zurich (CH)
- Corresponding author
- Kortemeyer, Gerd
- ORCID
- Kortemeyer, Gerd
- Venue
- TUGboat
- Publisher
- TeX Users Group
- Published
- 2023
- Volume
- 44
- Issue
- 3
- Pages
- 390-399
- ISSN
- 0896-3207 (ISSN-L)
- OpenAlex
- W4389027761
- Cited by
- 14
- References
- 0
- Citations by year
- 2026: 2 · 2025: 5 · 2024: 4 · 2023: 3
- FWCI
- 13.88
- Type
- journal article
- Language
- en
- Open access
- bronze
- doi.org
- Locations
- TUGboat (publishedVersion)
- arXiv (Cornell University) (submittedVersion, cc-by)
- arXiv (Cornell University) (cc-by)
- Topics
- Social Sciences › Human Factors and Ergonomics › Digital Accessibility for Disabilities
- Physical Sciences › Computer Science › Computational Theory and Mathematics › Mathematics, Computing, and Information Processing
- Health Sciences › Health Professions › Occupational Therapy › Assistive Technology in Communication and Mobility
- Keywords
- accessibility, LaTeX, mathematical formulas, screen readers, Braille display, assistive technology, jaw, NVDA, voice-over, visually impaired users
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
