Elderly Companion Avatar

The idea was to provide the “brain” with as many possible natural language interactions, or expanding the training data set for the machine learning algorithm. The project consisted of the development of a web-based user interface to facilitate companion to the elderly population. The application implemented the integration of different technologies such as JavaScript powered human-alike avatars with facial expressions, speech to text and text to speech engines, Google DialogFlow for language processing, a backend running Node.js on Linux, and a MySQL database.

K

GOOGLE DIALOGFLOW

K

GOOGLE TEXT TO SPEECH

K

GOOGLE SPEECH TO TEXT

K

CHATBOTS

K

JAVASCRIPT, NODE.JS, PHP, MYSQL

The Challenge

To build a new more affordable and user-friendly user interface to facilitate the access to an Elderly Companion software, and to nurture the “brain” of the robotic teacher assistant and the elderly companion robot.

The Solution

A responsive Web Application with a human-like interface, which accepted voice as its main data input. The application had user identification and sessions management developed with PHP. The core of the application was built with JavaScript (client and server-side), DialogFlow, SitePal avatars, and MySQL.

The Results

An application which was accessible from a desktop, tablet, or mobile device, featuring a human-like face, voice recognition, and delivery of information as text and speech. The audio commands were captured through the browser, converted to text, and sent to DialogFlow for language processing. Depending on the results from DialogFlow, different resources were activated in the backend. All the interactions with the avatar were used to feed the natural language conversational model of the system, and to collect relevant information from the user to better assist them.

Get In Touch

Ready to Chat?