This is a submission for the AssemblyAI Challenge : Sophisticated Speech-to-Text.

What I Built

VisualQuest is an interactive choose your own adventure AI application that allows you to create an immersive story from an uploaded image where your voice allows you to control the flow of the story.

Demo

You can access the app live here
The source code is available here

Journey

Universal-2 was incorporated into this application by having the user carry the story forward by making their decisions through their voice.
Each choice must be made through audio, where the Universal-2 model came into action, transcribing the detailed speech into textual response to feed it forward into the Llama model for creating the next segment of the story.

This submission only qualifies for the Universal-2 category