Peya Mowar
peyajm29 [at] cmu [dot] edu

A sketch of the head of Scotty, CMU's mascot scottish terrier. Below its collar is handwritten - 'My heart is in the work' and signed @peyajm29.

For a comprehensive and up-to-date list of my publications, please refer to Google Scholar.
Recently, I have been thinking about:

Accessibility Developer Tools

A long-standing challenge in accessible computing has been to get developers to produce the accessible UI code necessary for assistive technologies to work properly. With the increasing adoption of AI coding assistants, we investigate whether these tools can aid novice and untrained developers in writing accessible code by offering feedback and guidance.

A three-panel comic. 1: A sad engineer faces a pile of complexity (languages, toolchains, infra) leading to apps. 2: 'What People Think AI Will Do'—AI hides the complexity, making a non-engineer happy. 3: 'What Is Actually Going to Happen'—A new pile of AI-related complexity appears leaving the engineer even sadder.
CodeA11y: Making AI Coding Assistants Useful for Accessible Web Development
Peya Mowar, Yi-Hao Peng, Jason Wu, Aaron Steinfeld, Jeffrey Bigham
Under Submission, 2024

Tab to Autocomplete: The Effects of AI Coding Assistants on Web Accessibility
Peya Mowar, Yi-Hao Peng, Aaron Steinfeld, Jeffrey Bigham
Forthcoming at ASSETS (Posters Track), 2024
paper / video / slides

Accessibility in AI-Assisted Web Development
Peya Mowar
ACM International Web for All Conference (W4A) Doctoral Consortium, 2024
abstract / slides

Media Accessibility

Through mixed-methods research, we understand the challenges that print-impaired people face in accessing news media. We develop a document segmentation system for digitizing print newspapers by iteratively prototyping with 32 blind users. The system is now deployed by Microsoft as a Telegram chatbot for delivering daily news in screenreader-accessible formats.

A sketch of two hands holding an Indian newspaper, divided into various segments marked by different bounding boxes. These boxes represent the visualization output from an image segmentation system used for digitizing print newspapers.
Breaking the News Barrier: Towards Understanding News Consumption Practices among BVI Individuals in India
Peya Mowar, Meghna Gupta, Mohit Jain
Forthcoming at ASSETS, 2024
paper / video / slides

A qualitative study with BVI participants on their news consumption practices providing insights for designing accessible digital news platforms.

Towards Optimizing OCR for Accessibility
Peya Mowar, Tanuja Ganu, Saikat Guha
AVA: Accessibility, Vision, and Autonomy Meet (CVPR Workshop), 2022
paper

Identification of key visual cues that can significantly enhance the reading experience for print-impaired individuals through aural formats.


To chat, please email me or book a slot in my weekly office hours.
Modified template by Jon Barron
Last updated: September 2024