ML for Accessibility

28th October 2021, Friday


Recent advances in the fields of Artificial Intelligence and Machine Learning have opened up a host of new opportunities impacting life in unprecedented ways. In the modern era, access to technology has been crucial for even the most basic of daily activities. Hence, it is important to build inclusive and accessible technology that makes it possible for people with sensorimotor disabilities to take advantage of. The aim of this workshop is to present to the audience some of the exciting research activities happening at the intersection of Machine Learning and Accessibility in Google and beyond. The workshop is targeted towards Machine Learning practitioners who are eager to learn about and contribute to the development of accessible technologies.

1.Welcome note + workshop overview

2.Keynote Speech

3.Lookout: Computer vision to assist people with low vision or blindness

4.Lessons on developing HCI for blind and low vision subjects from extensive UXR studies

Vaishnav Kameshwaran

University of Michigan

5.Project Guideline: On-device machine learning to allow blind people to run independently using a phone, headphones and a guideline painted on the ground.

Xuan Yang

Short Break Session (15 mins)

6.Building robots for teaching children with special abilities

7.Building accessible user interfaces

8.Building accessibility tools: Touchless typing, lipreading and speech reconstruction based on lip movement

Rajiv Ratan Shah
Indraprastha Institute of Information Technology (IIIT) Delhi


Anirban Santara
Research Software Engineer @ Google Research India

Sohan Rudra
Pre-doctoral Researcher @ Google Research India

Gaurav Aggarwal
Staff Research Scientist @ Google Research India

Creating Accessible STEM Content

28th October 2021, Friday


Session Chair : M. Balakrishnan, IIT Delhi

Very few visually impaired students in India are able to pursue higher education in STEM disciplines. One of the key barriers has been inaccessible study material – whether it be diagrams or equations or tables. One approach to making diagrams accessible is through the tactile route which is definitely a preferred path for school textbooks. On the other hand a lot of content today is digital and may be required by only one or very few students enrolled in a certain course. This is especially true in higher education. There are a number of tools available – proprietary as well as open source that can be used to make contents accessible. The course would expose the participants to these tools. We would also give an introduction to RAVI – Reading Assistant for Visually Impaired, a comprehensive tool under development at IIT Delhi for making contents accessible.


Prof. Volker Sorge, University of Birmingham, UK

Neha Jadhav, ASSISTECH, IIT Delhi

Abhishek Baghel, Raised Lines Foundation

Who should attend

  • Visually impaired students attending high-school and college education in STEM subjects or their support persons
  • Special educators or others associated with NGOs, accessibility units
    in higher education Institutions and senior STEM teachers in high
    schools dealing with visually impaired students
Scroll Up