Talking Pumpkin

It's a pumpkin that doesn't want you to steal from its candy bowl! It'll have a regular conversation with you, and might even give you a piece of candy if you ask nicely

Click here to see it in action!

Contribution Notes

• This project uses Python 3.11.2
• Create a virtual environment and then run the following to get necessary packages: pip install -r requirements.txt
• Generate an OpenAI API Key, create a .env file, and add the following line to it: OPENAI_API_KEY = <insert your key>
• Generate a Resemble API Key, add the following line to .env: RESEMBLE_API_KEY = <insert your key>
• Update recording paths in main.py to match your recordings
• Enjoy!

Credits

• Arthur Bright for implementing text to speech and optimizing the speech to text code
• Luxman Sumanan and Catie Balasubramanian for giving insights on the mechanical design (which I should have paid more attention to)
• My parents for supporting this project via time, resources, and carving the pumpkin itself