A few years ago I took a tech support call from a guy who was visually impaired and had voice over turned on. It read incredibly fast, he had no problems following instructions, and was familiar enough with what we were doing to make the call easy for both of us.

I also used to work as a personal assistant for a guy in a wheelchair who used Dragon Speakeasy, I think that was back in 2005 - '06ish, and was surprised how well he was able to use his PC. He also had a book turning machine that reliably turned one page at a time.

This also reminds me of people who ride bikes using echo location.[1]

I wonder if it's possible to generate a sound stream from video, and what sort of resolution and colour depth could be encoded in to the sound.

This TED talk by Daniel Kish is good too.[2]

1. https://www.youtube.com/watch?v=Z_E3zxx2l9g 2. https://www.youtube.com/watch?v=CRA-asTuP_Y