The description of landscapes sounds interesting; the rest of the features—at least to me—sound a little condescending, and it would be better to make a product that does one thing well.


How will your product be better than the scene descriptions offered by apps like Seeing AI? How will you work out what details are relevant, and how will you provide an accurate description of a scene when the technology for analyzing pictures that Facebook, Apple, Google and Microsoft have, which has billions of images and enormous computing power cannot? How will a blind person have time to listen to scene descriptions when they should be using their ears to listen to more pertinent details in their environment that can't be captured with your camera because it's out of view or a sudden change?

Rather than assuming what blind people want, you need to do some research to investigate the feasability and challenges of any technology you want to develop.