Greeting AI

Inspiration

Nowadays, it is a common phenomenon that people feel tired of giving greetings to their friends or colleagues during holidays. A survey made by the American Psychological Association found that 38% of the people surveyed experienced stress during the holidays due to lack of time, the financial pressure of gift-giving, and others. Another survey made by the nation's largest grassroots mental health organization also found that 64% of people living with mental illnesses felt worse during this time. Furthermore, companies spend a lot of money on festival greetings in order to maintain a good connection with their customers. Our application, Greeting AI, delivers an ultimate solution that can solve the problem with a very low cost!

What it does

To Connect. To Share. To Greet. To Save Time & Money. Greeting AI is a Python desktop application that helps users and companies automatically send MMS message that contains the auto AI-generated greeting script and auto AI-generated greeting cards to the designated receivers. The only thing the sender will need to do is input the receiver's name, phone number, and a prompt about the festival such as "Christmas". Every receiver then will receive a unique human-like greeting script and greeting card based on our implemented AI and machine learning techniques; so that all of our friends, family members, colleagues, customers and etc. can actually feel the warmth without costing too much time and money for the sender.

How we built it

The GUI of the desktop application Greeting AI was built using Python Tkinter. Then we used OpenAI with text-davinci-002 model to generate the greeting scripts based on the user's prompt. Next, we applied DeepAI to generate greeting cards based on the same user's prompt. Lastly, we used Twilio to send the MMS messages that contain the auto AI-generated scripts and cards to the receiver list. The receiver list can be saved, loaded, and edited by using JSON from the sender's local repository.

Challenges we ran into

None of us knew how to implement a desktop application using Python. So we thought this is a good chance to learn the GUI framework in this hackathon. Indeed, the learning was easier than we thought, and we were able to implement more functionalities in this project such as the use of Twilio.

What we're proud of

Team management and work distribution were extremely organized during this hackathon. Every team member gets to show their unique talent in different stages of the software developmental process. More importantly, we managed to deliver the finished product with decent quality.

What's Next for Greeting AI

In the future, we want to add functionality that allows users to set up an automated greeting send-out function that can further reduce the heavy work from users. In this case, the sender does not need to worry about forgetting any festival in the future.