Introduction

In this aritcle, we will walk through the process of using Google Cloud Text-to-Speech API to convert text to speech in a Node.js application. This guide will cover setting up Google Cloud credentials, managing billing, preparing the source data, and running a script that converts phrases from JSON data into audio files.

Prerequisites

• Basic knowledge of JavaScript and Node.js.
• Node.js installed on your system.
• Google Cloud account.

Step 1: Setting up Google Cloud Console Credentials

1. Create a Google Cloud Project: Log in to the Google Cloud Console and create a new project.
2. Enable Text-to-Speech API: Navigate to the "API & Services" dashboard and enable the Text-to-Speech API for your project.
3. Create Credentials:
   • Go to the "Credentials" page.
   • Click "Create credentials" and select "Service account".
   • Follow the steps to create a service account.
   • Once created, click on your new service account and go to the "Keys" tab.
   • Add a new key and select JSON. The JSON file will be downloaded to your machine.

Step 2: Setting up Your Node.js Application

1. Initialize a Node.js Project: Create a new directory for your project and initialize it with npm init.
2. Install Dependencies: Install the necessary packages by running:

   bashCopy
   npm install @google-cloud/text-to-speech fs dotenv
   

3. Set Up Environment Variables:
   • Create a .env file in your project root.
   • Add the following line, replacing YOUR_CREDENTIALS with the content of the downloaded JSON file:

     GOOGLE_CREDENTIALS_CONTENT='YOUR_CREDENTIALS'
     

Step 3: Preparing the Source Data

Your source data should be in JSON format containing phrases you want to convert. Here is an example (/data/phrases.json):

jsonCopy
[
  {
    "id": "79546973-07ee-4b13-ae8e-2ebf07db7b2d",
    "phrase": "Potremmo vederci domani?",
    "translation": "Could we meet tomorrow?"
  }
  // Additional phrases...
]

Step 4: The Script

The provided script converts each phrase in the JSON file into an audio file using Google Cloud Text-to-Speech API.

jsCopy
const textToSpeech = require("@google-cloud/text-to-speech");
const fs = require("fs");
const util = require("util");
const path = require("path");
require("dotenv").config();

const phrases = require("../data/phrases.json");

const google_credentials_content = process.env.GOOGLE_CREDENTIALS_CONTENT;
const credentials = JSON.parse(google_credentials_content);
const client = new textToSpeech.TextToSpeechClient({ credentials });

async function convertTextToAudioFile(obj) {
  // console.log("convertTextToAudioFile:", obj.phrase);

  const request = {
    input: { text: obj.phrase },
    voice: { languageCode: "it-IT", ssmlGender: "FEMALE" },
    audioConfig: { audioEncoding: "MP3" },
  };

  const [response] = await client.synthesizeSpeech(request);

  // Specify the path where the audio should be saved
  const outputPath = path.join(
    __dirname,
    "../public/audio/",
    `${obj.id}.audio.mp3`
  );

  // Write the audio content to the file
  await fs.promises.writeFile(outputPath, response.audioContent, "binary");
}

async function convertAll() {
  for (const phraseObj of phrases) {
    await convertTextToAudioFile(phraseObj);
  }
}

convertAll();

Script Breakdown

• Load Dependencies: The script uses the @google-cloud/text-to-speech, fs, util, path, and dotenv modules.
• Load Phrases: It reads the phrases from phrases.json.
• Convert Text to Audio: For each phrase, the script sends a request to the Text-to-Speech API to generate audio and then saves it as an MP3 file.

Running the Script

Place the script in your project directory (e.g., as utils/process-audio.js) and run it using Node.js:

bashCopy
node utils/process-audio.js

Step 5: Managing Billing

Be aware of the billing for the Text-to-Speech API:

• Check Usage in Google Cloud Console: Regularly monitor your usage and charges in the Google Cloud Console.
• Set Alerts and Budgets: To prevent unexpected charges, set up billing alerts and budgets.
• Understand the Billing Cycle: Google Cloud TTS API follows a monthly billing cycle, and charges are based on the number of characters processed.

Conclusion

By following these steps, you can integrate Google Cloud Text-to-Speech into your Node.js application to create a dynamic and interactive voice experience. Remember to monitor your usage to align with your budget and regularly update your application as needed.