Dharanidharan01 / nexus

AI Storyteller is a web app powered by OpenAI's GPT-3.5-turbo-instruct engine, creating captivating narratives from user prompts and selected emotions.

AI Storyteller

ZIP file link

Zip_link

Presentation link

Canva

Overview

AI Storyteller is a web application that uses OpenAI's GPT-3.5-turbo-instruct engine to generate stories based on user prompts and selected emotions. The application also integrates HumeStreamClient for capturing facial emotions using a webcam and DALL·E for generating images related to the generated story.

Features

• Capture Emotion: Utilizes HumeStreamClient to capture facial emotion through a webcam.
• Generate Story: Takes user input prompts, modifies them based on selected emotions, and generates stories using OpenAI's GPT-3.5-turbo-instruct engine.
• Generate Images: Uses DALL·E to generate images corresponding to different sections of the generated story.
• Text-to-Speech: Utilizes OpenAI TTS to convert the generated story into audio for playback.

Installation

1. Clone the repository:

   git clone https://github.com/your-username/ai-storyteller.git

2. Install the required dependencies: pip install -r requirements.txt

3. Obtain API keys: OpenAI API key: OpenAI API HumeStream API key: Obtain from HumeStream (replace the placeholder in face.py)

Usage

Run the application: streamlit run storyteller.py

Dependencies

Streamlit OpenAI Hume OpenCV asyncio numpy

Configuration

Update API keys in storyteller.py and face.py. Customize the list of emotions in emotion_options in storyteller.py.