Samantha OS¹ (Her)

Samantha is an AI assistant inspired by the movie Her. This project is built to provide real-time voice interactions using the Realtime API and Chainlit. Samantha acts as an agent that calls various tools to handle user requests, such as querying stock prices, executing SQL commands, generating images, and creating Python scripts.

YouTube Demo

Watch a 2-minute demonstration of Samantha in action, where I showcase real-time voice interactions and various capabilities of the AI assistant.

Setup and Running

You can run Samantha either by setting up a virtual environment using uv or using Docker Compose. Note that setting up the environment variables is required in all cases.

Option 1: Using Virtual Environment

1. Clone the Repository

   git clone https://github.com/jesuscopado/samantha-os1.git
   cd samantha-os1
   

2. Set Up Virtual Environment

   • Install uv package manager: Installation Instructions
   • Create the virtual environment:

     uv sync
     

   • Activate the virtual environment:

     source .venv/bin/activate  # On Windows use `.venv\Scripts\activate`
     

3. Environment Variables

   • Create a .env file in the root directory by copying .env.example and updating it with your own keys.

4. Run the Application

   cd app
   chainlit run samantha.py
   

Option 2: Using Docker Compose

1. Environment Variables

   • Create a .env file in the root directory by copying .env.example and updating it with your own keys.

2. Build and Run with Docker Compose

   • Make sure Docker and Docker Compose are installed.
   • Run the following command:

     docker-compose up --d
     

Tools and Features

This project includes several powerful tools:

• Stock Price Queries: Using the yfinance package, Samantha can query the latest stock price information.
• Plotly Charts: Visualizations are created using Plotly to provide insights.
• Image Generation: Samantha can generate images using AI models through the Together API.
• Browser Interaction: Open web pages based on prompts.
• Internet Search: Samantha can perform internet searches via the Tavily API.
• LinkedIn Post Drafting: Create LinkedIn posts based on given topics using an AI model.
• Python Script Generation: Generate Python scripts on-demand based on user-provided topics.
• Python File Execution: Create and execute Python scripts directly from the assistant.
• 🆕 Database Interaction: Convert natural language to SQL and execute db commands with formatted results.
• 🆕 Email Drafting: Compose professional, personalized emails with appropriate subject lines and content.