Aify

Anubhav Singh

Anubhav Singh

Kolkata, West Bengal

0 0
  • 0 Collaborators

A single library including which you can automatically bring the power of AI into your website, making it more accessible to people with visual and cognitive disabilities. ...learn more

Project status: Published/In Market

Internet of Things, Artificial Intelligence

Intel Technologies
DevCloud, Intel Python

Code Samples [1]Links [1]

Overview / Usage

What is aify.js?

In the simplest terms, aify.js is a library which when embedded on your web page, can quickly add the power of AI to your website in the form of several features, without you having to invest in any backend or machine learning costs.

Features
  • Captioning for images on page
  • Label determination for page elements
  • Web page short summary
  • Voice UX
  • Browser Extension

Methodology / Approach

In the present date, people with visual impairment mostly use screen readers to access the internet. Some notable screen readers are the ChromeVOX which comes integrated with the Chrome browser and JAWS which is currently the most widely used screen reader, but for them to work properly the website must be coded for accessibility.

The core problem was identified as web developers not adding the required tags on the web pages while developing the websites. The task of adding these tags requires developers to make 100s of tiny changes on the websites, thus consuming their time and energy.

We provide the developers with a JavaScript library which they can include using a single line of code on their web page. This library automatically generates alternative text and updates the source code of the website while the user is browsing through it, without disrupting the user experience.

Our product relies on a set of APIs which perform the operation of generating alternative text for images and correctly labelling elements such as links and form fields on the web page. While the library packs in itself a number of features which are freely available, we decided to impose charges on the number of API calls being made which required significant work from the backend.

Technologies Used

PyTorch, Python, JavaScript, Google Cloud Platform, Intel AI DevCloud,

Repository

https://aifyjs.github.io/how-to/

Comments (0)