Watson Helper for the Visually Impaired

A simple web-app to describe images by voice to visually impaired users.

Instructions

• Download or clone this repository

• Put your service credentials in a .env file at the root of the project

# Visual Recognition
VR_API_KEY=
VR_VERSION=v3
VR_VERSION_DATE=2016-05-20

# Text to Speech
TTS_USERNAME=
TTS_PASSWORD=

• Install dependencies

npm install

• Run from command line

npm start

TODO

• A little bit more of CSS - make the ui usable by everyone (see here)
• Add Natural Language Understanding to describe the content of a website (main topic, authors...)
• Add Language Translation in the middle
• React all the things!
• Bonus: Tamper/Greasemonkey plugin (might reach the limit for Visual Recognition too quickly...)