Inspiration
We wanted to make healthcare more accessible, especially for elderly people and people who are not good with technology.
What it does
Our project is a demo for an AI healthcare assistant. It allows you to call a phone number and be connected to an AI that can help you keep track of your medications, post-op instructions, etc.
How we built it
We used Twilio Voice to receive voice calls and forward the voice data to Google App Engine. We then used a Flask web server to forward the data to the OpenAI Realtime API.
Challenges we ran into
We had trouble getting Twilio to successfully connect to the Google App Engine websocket. Thankfully, we were able to
Accomplishments that we're proud of
This is the first time we have made a program that you can call and interact with over the phone.
What we learned
We learned how to use the OpenAI Realtime API, as well as how to connect it to a phone number.
What's next for AI Healthcare Manager
Future development for this would be integrating it with healthcare providers so that it could get actual medical records, as well as adding a doctor portal so that doctors could add information if needed. Another feature will be the ability to have the AI send notes to your doctor, find in network specialists nearby, and to automatically start speaking in the user's preferred language.
Built With
- google-app-engine
- openai-realtime-api
- python
- twilio
Log in or sign up for Devpost to join the conversation.