WhatsApp Persona Mirror

WhatsApp Persona Mirror is a Streamlit application that analyzes WhatsApp chat history from JSON exports, generating personalized responses that imitate a selected participant's communication style and tone. Users can interact with the app to explore different conversational styles based on previous chats.

Features

• Role-Based Imitation: Select a chat participant’s role from the WhatsApp JSON file to generate responses in their unique style.
• Tone Customization: Choose from multiple conversation tones—Charming and Fun, Sarcastic, or Professional—to add personality to responses.
• Context-Aware Responses: Integrates relevant chat context using FAISS vector stores for more accurate, contextual replies.
• Streamlined Interface: Upload chat files, select parameters, and receive style-adapted responses on the go.

Installation

1. Clone the repository:

   (https://github.com/siddheshsonawane07/mimic_user_persona_using_mistral)
           

2. Set up a virtual environment:

   python3 -m venv env
   source env/bin/activate  # On Windows, use `env\Scripts\activate`
           

3. Install dependencies:

   pip install -r requirements.txt
           

4. Environment Variables:
   Create a .env file in the project root with your API key:

   MISTRAL_API_KEY=your_mistral_api_key
           

5. Run the Streamlit app:

   streamlit run app.py
           

Usage

1. Upload a WhatsApp Chat JSON File:
   • Export your chat from WhatsApp.
   • Upload this file on to this website to get the required format https://sidjsonifywhatsapptext.streamlit.app/.
2. Select Role and Tone:
   • Choose a participant's role from the dropdown list.
   • Pick a conversation tone (Charming and Fun, Sarcastic, Professional).
3. Start Chatting:
   • Type a message and observe responses tailored to your chosen style and tone.
   • Clear the chat history anytime using the Clear Chat button.

Project Structure

├── app.py                    # Main Streamlit application
├── README.md                 # Project documentation
└── .env.example              # Sample environment file
└── requirements.txt          # Requirements.txt file for downloading accurate dependencies

Dependencies

• Streamlit for app interface
• FAISS for vector-based similarity search
• Mistral AI for embedding and chat models
• tiktoken for token estimation

Troubleshooting

• Environment Variable Errors: Ensure MISTRAL_API_KEY is correctly set in your .env file.
• JSON Upload Errors: Only valid JSON files exported from WhatsApp are supported. If there’s an error, try re-exporting the JSON file.