Below are two minimal TypeScript examples using the Eleven v3 model.
- Text-to-Speech (Node SDK)
- Install: npm i @elevenlabs/elevenlabs-js
- Example:
import { ElevenLabsClient, play } from '@elevenlabs/elevenlabs-js';
import 'dotenv/config';
const client = new ElevenLabsClient({ apiKey: process.env.ELEVENLABS_API_KEY });
const voiceId = 'JBFqnCBsd6RMkjVDRZzb';
const audio = await client.textToSpeech.convert(voiceId, {
text: '[slowly] Back then... [chuckles] we had no phones. [whispers] Just dirt roads and [coughs] big dreams. [sad] Then it happened',
modelId: 'eleven_v3',
outputFormat: 'mp3_44100_128',
});
await play(audio);
Source for the TypeScript pattern and modelId usage: elevenlabs.io/v3 and the official JS SDK readme. (elevenlabs.io, github.com)
- Text-to-Dialogue (HTTP API via fetch)
import 'dotenv/config';
import { writeFileSync } from 'node:fs';
const res = await fetch('https://api.elevenlabs.io/v1/text-to-dialogue', {
method: 'POST',
headers: {
'xi-api-key': process.env.ELEVENLABS_API_KEY!,
'Content-Type': 'application/json',
},
body: JSON.stringify({
model_id: 'eleven_v3',
inputs: [
{ text: '[cheerfully] Hello, how are you?', voice_id: '9BWtsMINqrJLrRacOk9x' },
{ text: "[stuttering] I'm... I'm doing well, thank you", voice_id: 'IKne3meq5aSn9XLyUdCD' },
],
output_format: 'mp3_44100_128',
}),
});
const audio = Buffer.from(await res.arrayBuffer());
writeFileSync('dialogue.mp3', audio);
Endpoint and payload fields are from the Text-to-Dialogue API reference and quickstart. (elevenlabs.io)
Important notes
- Eleven v3 is in alpha, optimized for expressive, high-quality generations (not real-time). Consider generating multiple candidates and letting users pick the best. You can also target preview builds via model IDs like eleven_v3_preview_YYYY_MM_DD. (elevenlabs.io)
- Availability: Documentation is currently mixed. The v3 landing page shows “Build with the Eleven v3 API” and a TypeScript example, while other docs still say API access is not yet public and to contact sales. If your key lacks access, reach out to sales. (elevenlabs.io, help.elevenlabs.io)
If you want this adapted to your stack (ESM/CommonJS, Bun/Deno, streaming to a file or player), tell me your runtime and I’ll tailor it. (github.com)
Sources:
Eleven v3 (alpha) — The most expressive Text to Speech model
GitHub - elevenlabs/elevenlabs-js: The official JavaScript (Node) library for ElevenLabs Text to Speech.
Create dialogue | ElevenLabs Documentation
Models | ElevenLabs Documentation
Eleven v3 (alpha) — The most expressive Text to Speech model
What is Eleven v3 (Alpha)? – ElevenLabs
elevenlabs-js/reference.md at main · elevenlabs/elevenlabs-js · GitHub
feral vine
magic nest
umbral widget