Learnings from deploying a voice-based social platform for people with disability
- Karn Dubey ,
- Palash Gupta ,
- Rachna Shriwas ,
- Gayatri Gulvady ,
- Amit Sharma
ACM SIGCAS Conference on Computing & Sustainable Societies (COMPASS) |
Organized by ACM
For people with disability living in low-income neighborhoods, access to technology is compounded by inaccessible designs and relative isolation in poverty. To bring together this segment of population, an NGO in India built Enable Vaani, a voice-based social media platform designed for persons with disability living in rural areas without internet connectivity. This system has been in deployment since 2016 and has reached over 25,000 users. We present a mixed-methods analysis utilizing system logs and qualitative interviews, with the goal of understanding Enable Vaani’s impact. We find that posts related to employment, education and government programs are the most listened to and shared. Content uploaded by the Enable Vaani team is listened and bookmarked more per post than any user-generated content, indicating the importance of an active role of platform managers. Besides providing useful information, people reported that the platform serves as a medium for social support. Our analysis also finds areas of improvement: people skip a lot of posts, very few people share, and a large fraction of newcomers leave after the first month. In response to these shortcomings, we suggest strategies that can be useful for voice-based systems: active contribution of content, customized content feed, special strategies for encouraging newcomers, and design changes to make sharing easier.