NAO-Gemini/Llama Conversation Interface & Gesture imitation

The first project allows you to interact with Google's Gemini or Meta's LLama3 conversational AI through a NAO robot. The second will make the robot move by imitating your body movements (with an azure kinect DK camera) (still in developement)

The robots eyes will turn yellow when he detects your face, then you can start speaking and the eyes will turn green when recording your request.

Installation

Follow these steps to set up the environment and install the necessary dependencies.

Prerequisites

• Python 2.7 and Python 3.x
• paramiko
• naoqi
• speech_recognition
• google.generativeai
• pykinect_azure
• socket
• numpy

Setup

1. Clone the repository:

   git clone https://github.com/mil0sou/Nao-Conversation-Gestures
   cd NaoGeminiGestures/NaoGemini
   or 
   cd NaoGeminiGestures/NaoGestures

2. Install the dependencies:

   pip install numpy pykinect_azure llamaapi python-dotenv SpeechRecognition  
   pip install paramiko
   pip install -U google-generativeai

   download the 2.1.4 version of the SDK and the 32 bits python 2.7 msi file

   direct naoqi download link python 2.7 32 bits windows installer azure kinect body tracking sdk installer naoqi download page naoqi installation guide Microsoft Visual Package

3. Edit the code

   Edit each file to use your NAO IP, your computer paths (IN MOST FILES !), your google API key or llama API key Create a file named .env and write your keys in it :

   GOOGLE_API_KEY= #your keys here
   LLAMA_API_KEY=

Usage

Make sure NAO and your computer are connected to the same network.

To start the conversation interface between the NAO robot and Google's Gemini AI, or gesture imitation : run in either folder:

python3 main.py