Gemini API Express Server

This project is a simple Node.js Express server that provides two API endpoints for interacting with the Google Gemini API.

Endpoints

1. POST /analyze-audio: Accepts an audio file, analyzes it with a system prompt for context, and returns a structured JSON object containing the analysis (e.g., summary, sentiment, action items).

2. POST /generate-json: Accepts a text prompt and returns a structured JSON object based on the content of the prompt (e.g., extracting recipe details).

Prerequisites

• Node.js (v18 or later recommended)

• npm (comes with Node.js)

• A Google Gemini API Key. You can get one from Google AI Studio.

Setup and Installation

1. Clone the repository or save the files: Download server.js and package.json into a new project folder.

2. Create a .env file: In the root of your project directory, create a new file named .env and add your Gemini API key to it:

   GEMINI_API_KEY="YOUR_GEMINI_API_KEY_HERE"
   

3. Install dependencies: Open your terminal in the project directory and run:

   npm install
   

4. Start the server:

   npm start
   

   The server should now be running on http://localhost:3000.

How to Test the Endpoints

You can use a tool like curl or Postman to test the endpoints.

1. Test /analyze-audio

Find an audio file on your computer (e.g., an .mp3 or .wav file) and run the following curl command in your terminal, replacing path/to/your/audio.mp3 with the actual path to your file.

curl -X POST http://localhost:3000/analyze-audio \
  -F "audio=@path/to/your/audio.mp3"

Expected Response:

You should receive a JSON object similar to this:

{
    "summary": "The customer is calling to report an issue with their recent order where they received the wrong item.",
    "sentiment": "negative",
    "action_items": [
        "Apologize to the customer for the inconvenience.",
        "Initiate a return for the incorrect item.",
        "Ship the correct item with expedited shipping."
    ]
}

2. Test /generate-json

Run the following curl command in your terminal. The System Prompts are not calibrated, make sure you fix it before testing, or the endpoint will output some other random format.

curl -X POST http://localhost:3000/generate-json \
  -H "Content-Type: application/json" \
  -d '{
        "prompt": "Create a simple recipe for chocolate chip cookies. It should make 24 cookies and include flour, sugar, butter, and chocolate chips."
      }'

Expected Response:

You should receive a random JSON object structured like this:

{
    "recipeName": "Simple Chocolate Chip Cookies",
    "ingredients": [
        "1 cup all-purpose flour",
        "1/2 cup granulated sugar",
        "1/2 cup unsalted butter, softened",
        "1 cup chocolate chips"
    ],
    "servings": 24
}