Audio Generation (TTS)

Convert text to speech. The response is raw binary audio data.

Basic Usage

const response = await client.audio.create({
  model: "openai/tts",
  input: "Hello from Lunos!",
  voice: "alloy",
  response_format: "mp3",
  speed: 1.0,
});

// Write audio to file (Node.js)
import { writeFileSync } from "fs";
writeFileSync("output.mp3", Buffer.from(await response.arrayBuffer()));

Parameters

Parameter	Type	Description
model	string	TTS model identifier
input	string	Text to convert to speech
voice	string	Voice selection
response_format	'mp3' | 'opus' | 'aac' | 'flac' | 'wav'	Audio format
speed	number	Speed multiplier (0.25–4.0)

Available Voices

Voice	Description
alloy	Neutral, balanced
echo	Warm, conversational
fable	Expressive, storytelling
onyx	Deep, authoritative
nova	Friendly, upbeat
shimmer	Clear, professional

Streaming to Client

In a web server context, pipe the audio response directly:

// Express.js example
app.get("/speak", async (req, res) => {
  const response = await client.audio.create({
    model: "openai/tts",
    input: req.query.text,
    voice: "nova",
    response_format: "mp3",
  });

  res.setHeader("Content-Type", "audio/mpeg");
  const buffer = Buffer.from(await response.arrayBuffer());
  res.send(buffer);
});