The AI-Powered Image Storyteller is an innovative application that leverages cutting-edge AI models from HuggingFace and OpenAI to turn images into captivating stories. The primary objective is to provide users with a narrative generated from an uploaded image.
-
User Uploads an Image: The user starts by uploading an image of their choice.
-
Image to Text Conversion: We employ an HuggingFace model, specifically "salesforce/blip-image-captioning-base," to convert the image into descriptive text.
-
Langchain Integration: Using Langchain, we make use of an OpenAI model, specifically GPT-3.5, with a prompt template to craft a compelling story based on the image.
-
Story to Speech: The response story is then converted back into spoken words using HuggingFace's speech-to-text model.
-
User Interface: The user interacts with the application through a user-friendly interface powered by the Streamlit library.
To configure this application, make sure to add an .env file for storing the necessary API keys:
# .env file
HUGGINGFACE_API_KEY=your_huggingface_api_key
OPENAI_API_KEY=your_openai_api_key
To experience this AI-powered storytelling application, simply run the following command:
streamlit run app.py
This project is a testament to the fusion of cutting-edge AI technology with a user-friendly interface, making image-based storytelling accessible and engaging for all.
Happy storytelling! 📸📚🎙️