EmailBot AI

Inspiration

This project was heavily inspired by the current situation in Burkina Faso, a west-African country where a massive portion of the population lacks access to internet. This severely limits education for students in the country, and we believe that education is a right that all should have access to. This is why we thought of TextBot AI, a chatbot that does not require internet to access. While most phone services utilize internet, SMS-based text messages can send without internet, using satellite data instead. On this basis, we came up with the idea of a chatbot connected to a phone number that you could text any educational-based question, and it would text an answer right back to you within seconds, almost like texting a friend that's always online. We could theoretically set up SMS-based stations around underprivileged school districts in Burkina Faso where students could ask any education-based questions on their mind and receive answers, allowing them to have access to a vast gallery of knowledge.

Original idea and the birth of EmailBot AI

While we originally wanted to use text messages, it's extremely hard to get ahold of a phone number within days, especially when companies hesitate to provide programable phone numbers without requesting some sort of payment. We had a budget of $0 for this project, and if we wanted to show you exactly what we wanted our project to do, we would need something different in order to do so. This is where we came up with the idea of using emails instead. While emails use internet, they allow for the same thing as text messages: communication. This is where EmailBot AI was born, a chatbot that will answer any of your questions if you just shoot it a quick email.

How you can use it

If you are willing to try out our final product, you must follow the directions given carefully. These are explained throughout this blurb, the video attached, and the link to our Google Colab.

1. Watch the video to gain a brief understanding of how the project works
2. Open Google Colab, and follow the directions there. It is important to note that you may be subject to high wait times as you load the coding boxes, with the second box taking up to 20 minutes depending on the speed of your computer. This is because using our chatbot requires you to locally download a model that we found on Hugging Face.
3. Send an email to [email protected], asking whatever educational-based question you may like. Please make sure to keep these school appropriate. As long as you keep the code running in the background on Google Colab, you should recieve a reply within 1-2 minutes. This can be slower or faster depending on the speed of your computer.
4. If you have any questions, please make sure that you properly watched the video, read this blurb, and read the instructions on Google Colab. If you seem to have trouble accessing something, do not hesitate to reach out to us at [email protected] or [email protected].

Built With

• artificialintelligence
• colab
• huggingface
• python