README.md

metadata

title: Autism_QA
app_file: multi_page_gradio_demo.py
sdk: gradio
sdk_version: 5.40.0

Wisal: Autism AI Assistant

Wisal is an AI-powered assistant specifically designed to help individuals with autism with their communication needs. The application provides both text and audio responses to make communication more accessible.

Features

• Text and voice input/output
• Document Q&A capabilities
• Live audio chat
• Customizable voice responses
• User-specific document handling

Prerequisites

Before you begin, ensure you have the following installed:

• Python 3.8 or higher
• pip (Python package installer)
• Conda (optional, but recommended for environment management)

Setup Instructions

1. Clone the Repository

git clone <repository-url>
cd autim-qa-mine

2. Environment Configuration

Copy the example environment file and fill in your API keys:

cp .env.example .env

Then edit the .env file and add your API keys:

• Google Gemini API Key
• SiliconFlow API Key
• Other optional API keys as needed

3. Running the Application

We provide a convenient script to run the application with different environment options:

./run.sh

The script will present you with three options:

1. Use an existing conda environment
2. Create a new conda environment
3. Install requirements with the current Python environment

Manual Installation (Alternative)

If you prefer to set up the environment manually:

1. Create a virtual environment (recommended):

   python -m venv wisal-env
   source wisal-env/bin/activate  # On Windows: wisal-env\Scripts\activate
   

2. Install dependencies:

   pip install -r requirements.txt
   

3. Run the application:

   python main.py
   

Usage

Once the application is running, it will be accessible at http://localhost:8080 in your web browser.

Basic Usage

1. Type your question in the text box or record audio using the microphone
2. Select your preferred voice for the response
3. Click "Send Message" or press Enter
4. The AI will respond with both text and audio

Advanced Features

• Document Q&A: Upload documents (PDF, DOCX, TXT) to ask questions about their content
• Live Chat: Use the WebRTC feature for real-time audio conversation
• User-Specific Documents: Store and query personal documents

Configuration

The application can be configured through the config.yaml file for model settings and other parameters.

Troubleshooting

If you encounter issues:

1. Ensure all API keys in .env are correct and active
2. Check that all dependencies are installed: pip install -r requirements.txt
3. Verify the application is using the correct Python environment

For further assistance, please check the logs in the logs/ directory.