Inworld TTS (Realtime TTS-2, TTS-2 Flash)

Overview

Best for: Cost-sensitive, high-volume voice agents and games that still want cloning and timestamps.

At a glance

Figures are TTS-2 (P90 server TTFB 100 ms, $25/1M On-Demand). TTS-2 Flash: 20 ms P90, $15/1M, ignores steering. Languages: 15 production quality per release notes; docs claim 200+. Concurrency 5 On-Demand (Creator 10, Builder 50). Free start includes up to 70 minutes. Zero data retention supported.

Features

• input streaming (bidirectional WebSocket)
• multiple contexts per connection
• autoMode sentence buffering
• word timestamps, phonemes and visemes (TTS-2)
• voice cloning
• voice design
• zero data retention
• OpenAI-compatible POST /v1/audio/speech (2026-09-11)

Pricing

Free tier: On-Demand: up to 70 minutes of TTS

Source: inworld.ai

Setup

1. Get an API key (Basic credential) from the Inworld portal.
2. Connect to wss://api.inworld.ai/tts/v1/voice:streamBidirectional with Authorization: Basic <key>.
3. Send create (voice_id, model_id, audio_config), then send_text messages, then flush_context / close_context.
4. Decode result.audioChunk.audioContent (base64).

Endpoint

wss://api.inworld.ai/tts/v1/voice:streamBidirectional

Authentication

Authorization: Basic <INWORLD_API_KEY> (the key from the portal is already Base64)

// npm i ws  - message shapes follow Inworld's official example_websocket.js
import WebSocket from "ws";
import fs from "fs";

const ws = new WebSocket("wss://api.inworld.ai/tts/v1/voice:streamBidirectional", {
  headers: { Authorization: `Basic ${process.env.INWORLD_API_KEY}` },
});
const out = fs.createWriteStream("out_24k_s16le.pcm");
const ctx = "turn-1";

ws.on("open", () => {
  ws.send(JSON.stringify({ context_id: ctx, create: {
    voice_id: "Ashley", model_id: "inworld-tts-2-flash",
    audio_config: { audio_encoding: "PCM", sample_rate_hertz: 24000 } } }));
  for (const t of ["Hello there. ", "Streaming text ", "from an LLM."])
    ws.send(JSON.stringify({ context_id: ctx, send_text: { text: t } }));
  ws.send(JSON.stringify({ context_id: ctx, close_context: {} }));
});
ws.on("message", (raw) => {
  const r = JSON.parse(raw.toString()).result;
  const b64 = r?.audioChunk?.audioContent || r?.audioContent;
  if (b64) out.write(Buffer.from(b64, "base64"));
  if (r?.contextClosed) ws.close();
});
ws.on("close", () => out.end());

Written from the current docs. Check the vendor's SDK version before you ship.

Warnings

Plus 12 warnings that apply to all text-to-speech, streaming APIs. See category warnings.

Limits

• Concurrent requests: On-Demand 5, Creator 10, Builder 50, Developer 150, Growth 500
• 2,000 UTF-16 code units per send_text message
• Socket closes after 10 min of inactivity across all contexts
• Sync 2,000 chars, HTTP streaming 4,000 chars, async 100,000 chars per request

Models and products

Docs and sources

Docs

• TTS models
• WebSocket API reference
• Release notes
• Official WebSocket example

Sources used

• inworld.ai/pricing
• docs.inworld.ai/tts/tts-models.md
• docs.inworld.ai/release-notes/tts
• docs.inworld.ai/api-reference/ttsAPI/texttospeech/synthesize-speech-websocket.md
• docs.inworld.ai/tts/synthesize-speech.md
• github.com/inworld-ai/inworld-api-examples/blob/main/tts/js/example_websocket.js