Text-to-Voice Conversion: OpenAI-Speech

Introduction

OpenAI-Speech is a Node.js library designed to transform textual input into realistic voice outputs using advanced text-to-speech technologies from providers like OpenAI and Google. This module is built to be easily integrated into various projects needing spoken output capabilities.

Setup Instructions

Installation Process

1. First, ensure Node.js is installed on your system.

2. To install the OpenAI-Speech module, run the following command:

npm install openai-speech

How to Use

The module offers two styles of interaction – a functional interface and an object-oriented interface:

• Functional Interface: Ideal for applications requiring a single text-to-speech configuration across the entire application.
• Object-Oriented Interface: Best for applications needing different configurations or multiple service providers.

Functional Interface Usage

This simpler, singleton-based interface suits applications using a uniform configuration:

const openaiSpeech = require('openai-speech');

// Example text
const text = "Experience OpenAI-Speech in action.";

// Initialization with API key
openaiSpeech.initialize(process.env.API_KEY);

// Convert text to speech
openaiSpeech.quickSpeech(text)
    .then(audio => {
        // Work with the audio output
    })
    .catch(error => {
        console.error("Error:", error);
    });

Object-Oriented Interface Usage

This interface allows for multiple configurations and providers within the same application:

const { OpenAISpeech, createProvider } = require('openai-speech');

// Repeating a text to extend its length
const longText = "Experience OpenAI-Speech in action. ".repeat(50);

// Initializing with a specific provider
const speechService = new OpenAISpeech(createProvider("Google", process.env.GOOGLE_CREDENTIALS));

// Convert longer text to speech
speechService.extendedSpeech(longText)
    .then(audio => {
        // Manage the audio output
    })
    .catch(error => {
        console.error("Error:", error);
    });

Configuring Providers

Use the createProvider factory to configure different providers as needed. If additional configuration is necessary, direct imports from openai-speech/providers may be required.

Command Line Interface (CLI) Usage

For direct command line use, the following command converts text from a file to speech:

cat your-text-file.txt | ./bin/openai-speech-cli --provider OpenAI > output.mp3

API Overview

Functions available:

• quickSpeech(text: string, options = {}): Promise<Buffer> – Converts text into speech quickly.
• extendedSpeech(text: string, options = {}): Promise<Buffer> – Handles longer text by breaking it into manageable parts.

Example Usage

For an example setup, see the main.js in the project repository. Run it using your API key:

OPENAI_API_KEY="your-api-key" node main.js --provider OpenAI

Alternatively, you can store and use your API key from an environment file:

echo 'OPENAI_API_KEY="your-api-key"' > .env
source .env
node main.js --provider OpenAI

Support and Enhancements

For any problems or improvement suggestions, please create an issue on the GitHub repository.

Licensing

OpenAI-Speech is made available under the MIT License, supporting free usage and distribution with appropriate credit.