Inspiration

The vocabulary learning system in the United States is seriously flawed. Students around the country are repeatedly quizzed on sets of vocabulary words that they tend to learn the night before. However, they soon forget the words after the quiz as they never learn the words in-context; they never associated the words with images, people, and stories. The true way to learn and remember vocabulary words is to read books, or use our project

What it does

This project takes in a set of vocabulary words and generates a story using the words such that the words are able to be learned given context clues. A new story is created every day and is emailed to a list of email addresses of people who want to receive the story.

How we built it

We used Python to program all of it, specifically the openai, email, time, and schedule libraries. First, we set up the schedule so that an email would be sent every day at a certain time. Next, we wrote a script to access the openai API so that we could ask it questions while running the script. We then changed the script so that it had a prewritten prompt rather than user input. We finally combined the two and used the scheduling script to send out the prewritten prompt response every day.

Challenges we ran into

Initially, the program was unusable because every night the prompt would stop running as the computer went to sleep. However, we implemented a condition that was always true so that even if the computer fell asleep the script would still run.

Accomplishments that we're proud of

We successfully used artificial intelligence to help improve society.

What we learned

We learned how to use the openai, email, time, and schedule libraries. We also learned how to work as a team and solve different components of a larger problem, as we had to break this project down into two sections-- the emailing and the AI response.

What's next for Vocabulary Story Generator

An improvement would be to make a website where a user could sign up to be emailed a story using vocabulary words of their choice. Currently, the programmer has to manually input the email addresses to be emailed as well as the specific vocabulary words.

Built With

  • email
  • openai
  • python
  • schedule
  • smptlib
Share this project:

Updates