VoiceAI SDK for TypeScript

The official TypeScript SDK for SLNG Voice AI. Use one client for text-to-speech, speech-to-text, and model discovery across SLNG-hosted and provider-backed speech models.

Install

npm install voiceai-sdk

Node 18 or newer is required.

API Key

Create an API key in the SLNG dashboard, then set it in your environment:

export SLNG_API_KEY="zpka_..."

import Slng from 'voiceai-sdk';

const client = new Slng();

You can also pass apiKey directly:

const client = new Slng({ apiKey: process.env.SLNG_API_KEY });

Text To Speech

Generate audio from text and save it locally:

import { writeFile } from 'node:fs/promises';
import Slng from 'voiceai-sdk';

const client = new Slng();

const response = await client.textToSpeech.create('slng/deepgram/aura:2-en', {
  text: 'Hello from SLNG.',
  voice: 'aura-2-thalia-en',
});

await writeFile('hello.wav', Buffer.from(await response.arrayBuffer()));

Use region or 'world-part' when you need explicit routing:

await client.textToSpeech.create('slng/deepgram/aura:2-en', {
  text: 'Hello from Europe.',
  voice: 'aura-2-thalia-en',
  region: 'eu-north-1',
});

Speech To Text

Transcribe a local audio file:

import fs from 'node:fs';
import Slng from 'voiceai-sdk';

const client = new Slng();

const transcript = await client.speechToText.create('slng/deepgram/nova:3-en', {
  audio: fs.createReadStream('meeting.wav'),
});

console.log(transcript.text);

File uploads accept File, Response, fs.ReadStream, or the SDK toFile helper.

Discover Models

The SDK ships with a static catalog snapshot, so agents and scripts can choose valid models without credentials or network calls.

import { getModel, listModels } from 'voiceai-sdk';

const ttsModels = listModels({ service: 'tts', language: 'en' });
const sttModels = listModels({ service: 'stt' });
const aura = getModel('slng/deepgram/aura:2-en');

console.log(ttsModels.map((model) => model.id));
console.log(aura?.deployments);

Model IDs from the catalog can be passed directly to client.textToSpeech.create() or client.speechToText.create().

Discover Voices

Use listVoices() to find voice IDs for a TTS model:

import { listVoices } from 'voiceai-sdk';

const voices = listVoices({
  model: 'slng/deepgram/aura:2-en',
  language: 'en',
});

console.log(voices.map((voice) => `${voice.name}: ${voice.voiceId}`));

listVoices() returns an empty array when a model does not have a cataloged voice list.

Useful voice references:

• Voices in the SLNG docs
• Deepgram Aura voices
• Cartesia Sonic 3 voices
• Kugel voices
• Murf Falcon voices
• Orpheus voices
• Rime Arcana voices
• Sarvam Bulbul voices
• Soniox TTS voices
• Full voice catalogs and samples are available in the SLNG dashboard.

Streaming

For low-latency WebSocket sessions, use the streaming client:

import { streaming } from 'voiceai-sdk';

const client = new streaming.StreamingClient({
  apiKey: process.env.SLNG_API_KEY!,
});

const session = await client.connectTts('slng/deepgram/aura:2-en');
session.send({
  type: 'init',
  model: 'slng/deepgram/aura:2-en',
  voice: 'aura-2-thalia-en',
});
session.send({ type: 'text', text: 'Streaming text to speech.' });
session.send({ type: 'flush' });

for await (const message of session) {
  console.log(message);
}

Reference

• Full API surface: api.md
• General docs: docs.slng.ai
• Models by language: docs.slng.ai/models/by-language
• Source: github.com/slng-ai/voiceai-sdk