Welcome to the Walkie-Talkie: A tool aimed at educating, engaging and empowering our elderly while complementing our healthcare services.
Walkie-Talkie is an AI-driven chatbot designed to assist elderly users with daily engagement, cognitive stimulation, and health-related support. By integrating conversational AI with healthcare guidance, the chatbot enhances the well-being of seniors while reducing the burden on caregivers and medical professionals. The chatbot also conducts sentiment analysis on user interactions based on the National Healthcare Group's 3E5P framework to provide healthcare practioners with information on patients' state of change. This enables more targetted and effective healthcare outcomes in the long run.
The chatbot utilises Retreival-Augmented-Generation (RAG) to search through provided MOH documents and provide users with accurate and up-to-date information. This is powered through the Intersystem's IRIS database, which allow for efficient storage and similarity lookup of these documents once vectorised
Please see the video below for further information:
Clone this repo via
git clone https://github.com/Nsohko/Walkie-Talkie.git
cd Walkie-Talkie
Note: You will need an OpenAI API key with credits to use this app.
Ensure that conda, node.js and yarn are installed.
Create a new conda env, and activate it:
conda create --name walkietalkie python=3.10
conda activate walkietalkie
In the project root, create .env file with your openAI API Key.
OPENAI_API_KEY="YOUR-KEY-HERE"
Install IRIS Community Edtion in a container. This will be your SQL database server. More information can be found here
docker run -d --name iris-comm -p 1972:1972 -p 52773:52773 -e IRIS_PASSWORD=demo -e IRIS_USERNAME=demo intersystemsdc/iris-community:latest
From the project root, enter the following commands:
cd backend
pip install -r requirements.txt
pip install ./install/intersystems_irispython-5.0.1-8026-cp38.cp39.cp310.cp311.cp312-cp38.cp39.cp310.cp311.cp312-win_amd64.whl
flask run
This will start up the backend. Note: The above is assuming this is being run on windows. For information on other Operating System, please see here
From the project root, enter the following commands
cd frontend
yarn install
yarn start
This will start the frontend
The service will now be available at http://localhost:3000/