Piper TTS Server

This Node.js application is a Text-to-Speech (TTS) server using Piper, which converts text input into spoken audio.

Requirements

Mac and Window (Ubuntu)

• Node.js
• Npm

Installation

1. Clone the repository.

   git clone https://github.com/flukexp/PiperTTS-API-Wrapper.git

2. Navigate to the project directory.

   cd PiperTTS-API-Wrapper

3. Install dependencies, Piper TTS, Piper sample voices and Start piper server:

   ./piper_installer.sh
   

Endpoints

Base URL

http://localhost:5001/

1. GET /

Description:
Returns basic information about the server.

2. POST /tts

Request Body: The request body should be a JSON object with the following properties:

• text (string, required): The text to be converted into speech.
• voice (string, optional): The voice model to be used for speech synthesis. If not provided, the default voice model will be used.

Response: audio/wav

3. GET /tts?text=<text>&voice=<voice>

Request Body: The request body should be a JSON object with the following properties:

• text (string, required): The text to be converted into speech.
• voice (string, optional): The voice model to be used for speech synthesis. If not provided, the default voice model will be used.

Response: audio/wav

4. GET /voices

Description: The response will be a JSON array containing the names of available voice models.