LiveAPI

Currently testing some LiveApi Stuff

Running the LiveTest Application

Prerequisites

• Python 3.8 or higher
• Gemini API key (get one from Google AI Studio)
• Working microphone
• Webcam (for camera mode)
• Speakers or headphones

Installation Steps

1. Clone the repository:

git clone https://github.com/Toltally-suck-at-code/LiveAPI.git
cd LiveAPI

2. Create and activate a virtual environment:

For Mac/Linux:

# Create virtual environment
python -m venv venv

# Activate virtual environment
source venv/bin/activate

For Windows (Command Prompt):

# Create virtual environment
python -m venv venv

# Activate virtual environment
venv\Scripts\activate.bat

For Windows (PowerShell):

# Create virtual environment
python -m venv venv

# Activate virtual environment
.\venv\Scripts\Activate.ps1

3. Install system dependencies:

For macOS:

# Install PortAudio using Homebrew
brew install portaudio

For Ubuntu/Debian Linux:

sudo apt-get install python3-pyaudio portaudio19-dev

For Windows:

# No additional system dependencies required

4. Install Python dependencies:

pip install -r requirements.txt

5. Set up your Gemini API key:

For Mac/Linux:

export GEMINI_API_KEY='your-api-key-here'

For Windows (Command Prompt):

set GEMINI_API_KEY=your-api-key-here

For Windows (PowerShell):

$env:GEMINI_API_KEY='your-api-key-here'

Running the Application

python /app.py

You can run the application in three different modes: (BACKUP MODE!)

1. Camera mode (default):

python Old/LiveTest.py --mode camera

2. Screen capture mode:

python Old/LiveTest.py --mode screen

3. Audio only mode:

python Old/LiveTest.py --mode none

Features

• Real-time audio interaction with Gemini AI
• Optional video streaming from camera or screen
• Push-to-talk functionality
• Voice responses from the AI

Troubleshooting

• If you encounter audio issues, check your microphone and speaker settings
• Make sure your webcam is working if using camera mode
• Verify that your Gemini API key is correctly set in the environment variables
• If you get dependency errors, try running pip install -r requirements.txt again
• If you get "command not found" errors, make sure your virtual environment is activated
• To deactivate the virtual environment when you're done, simply type deactivate in your terminal
• For macOS users: If you get PortAudio errors, make sure you've installed it using brew install portaudio
• For Linux users: If you get PortAudio errors, make sure you've installed the required packages using apt-get

Push-to-Talk Feature

Overview

This app now includes a push-to-talk (PTT) feature to prevent the AI from hearing its own responses through your laptop speakers.

How to Use

Mouse/Touch Control

• Click and hold the microphone button in the UI to activate voice input
• Release the button to stop transmitting audio

Keyboard Control

• Press and hold Space bar to activate voice input
• Release Space bar to stop transmitting audio

Visual Indicators

1. Microphone Button States:

   • Gray = Idle (not transmitting)
   • Green with pulse effect = Active (transmitting audio)
   • Faded/Disabled = No live session running

2. Listening Indicator:

   • A red "Listening..." indicator appears in the bottom-right of the video feed when PTT is active

Important Notes

• The microphone will only transmit audio while PTT is active
• PTT is automatically disabled when:
  • No live session is running
  • The browser window loses focus
  • You navigate away from the page
• The feature works on both desktop and mobile devices

Benefits

• Prevents audio feedback loops
• Gives you complete control over when to speak
• Reduces background noise transmission
• Saves bandwidth by only sending audio when needed