Inworld AI TTS

The InworldTTSService provides ultra-realistic, context-aware speech synthesis. It is designed for interactive characters and gaming experiences, offering precise voice control and human-like speech quality.

Installation

To use Inworld AI, install the required dependencies:

pip install "piopiy-ai[inworld]"

Prerequisites

• An Inworld AI API key (Get yours here).
• Set your API key in your environment:

  export INWORLD_API_KEY="your_api_key_here"

Configuration

InworldTTSService (WebSocket) Parameters

Parameter	Type	Default	Description
api_key	str	Required	Your Inworld API key.
voice_id	str	"Ashley"	ID of the voice to use.
model	str	"inworld-tts-1"	ID of the model to use.
params	InputParams	None	Configuration parameters (see below).

InputParams Options

Option	Type	Default	Description
temperature	float	None	Randomness in speech synthesis.
speaking_rate	float	None	Speech speed multiplier.
apply_text_normalization	str	None	Whether to pre-process text.

Usage

Basic Setup

import os
from piopiy.services.inworld.tts import InworldTTSService

tts = InworldTTSService(
    api_key=os.getenv("INWORLD_API_KEY"),
    voice_id="Ashley"
)

With Custom Settings

import os
from piopiy.services.inworld.tts import InworldTTSService

tts = InworldTTSService(
    api_key=os.getenv("INWORLD_API_KEY"),
    voice_id="Ashley",
    params=InworldTTSService.InputParams(
        temperature=0.8,
        speaking_rate=1.1
    )
)

Notes

• Real-time Interaction: The WebSocket service (InworldTTSService) is optimized for low-latency, bidirectional communication, making it ideal for interactive AI agents.
• Word Timestamps: This service automatically provides word-level timestamps, enabling precise synchronization of lip-sync and subtitles.