Will-E | Devpost

He's here.
He's there.
He's everywhere.

Inspiration

We have used AI before and wanted to see in what other interesting ways it could be utilized.

What it does

This program is capable of processing prompts via text or speech. So if you have a question that needs answering or a thought you would like to expand on, Will-E can assist in most languages. By starting a request with "Ok..." (Only English for now), it can also:

Create a folder on your Desktop
Open up your default browser
Evaluate storage usage on your desktop
Analyze most taxing program running on your machine
Pull up a song on YouTube
Search for things on Google

How we built it

100% of the code is Python
OpenAI is used to respond to the prompts
OpenAI handles speech-to-text
ElevenLabs handles text-to-speech (Will-E's voice)
Illustrator to create a spritesheet (Will-E designs)
Github for version control
Notepad++ and VSCode for editing

Challenges we ran into

Scrollbar wouldn't appear when it should've
Threads were hard to implement (UI updates, TTS, STT, Chat)
API keys expiring

Accomplishments that we're proud of

We're proud of the fact that we were able to finish this project. The task seemed overwhelming but we persevered. Practically every feature and capability that was imagined was implemented.

What we learned

We learned how to use two different AI APIs, work with text and speech synthesizers, a lot of the python libraries used, and how to create GUI in the overlay.