Skip to main content
The official TypeScript SDK is published as @supertone/supertone on npm. Source: supertone-inc/supertone-ts.

​
At a glance

Package@supertone/supertone on npm
Reposupertone-inc/supertone-ts
LanguagesTypeScript 5+ / Node 18+
Module formatsESM + CommonJS
Authnew Supertone({ apiKey: ... })
Sync APIn/a
Async APIAll methods (Promises)
StreamingReadableStream reader
Auto-chunk long text✅ 300 chars, sequential
Custom retries✅ via retryConfig
HTTP backendfetch

​
Installation

npm add @supertone/supertone
The package is published with both ESM and CommonJS entry points and ships with TypeScript types. Node 18+ is required (for the global fetch and ReadableStream). Works in Bun and Deno; not designed for the browser (your API key should never live client-side).

​
Set your API key

export SUPERTONE_API_KEY="Kp9mZ3xQ7v..."

import { Supertone } from "@supertone/supertone";

const client = new Supertone({ apiKey: process.env.SUPERTONE_API_KEY });
Voice IDs are not environment variables — they change per use case, so keep them as plain strings in your code (or pass them from your request payload).

​
Generate speech

import { Supertone } from "@supertone/supertone";
import * as fs from "node:fs";

const VOICE_ID = "20160a4c5ba38967330c84"; // replace with your voice ID

const client = new Supertone({ apiKey: process.env.SUPERTONE_API_KEY });

const response = await client.textToSpeech.createSpeech({
  voiceId: VOICE_ID,
  apiConvertTextToSpeechUsingCharacterRequest: {
    text: "Hello from the TypeScript SDK.",
    language: "en",
    model: "sona_speech_1",
    outputFormat: "wav",
  },
});

if (response.result instanceof Uint8Array) {
  fs.writeFileSync("speech.wav", response.result);
} else if (response.result && "getReader" in response.result) {
  const reader = (response.result as ReadableStream<Uint8Array>).getReader();
  const chunks: Uint8Array[] = [];
  while (true) {
    const { done, value } = await reader.read();
    if (done) break;
    if (value) chunks.push(value);
  }
  fs.writeFileSync("speech.wav", Buffer.concat(chunks));
}
Every method is async and returns a Promise. There is no sync alternative.

​
Stream speech

import { Supertone } from "@supertone/supertone";
import * as fs from "node:fs";

const VOICE_ID = "20160a4c5ba38967330c84"; // replace with your voice ID

const client = new Supertone({ apiKey: process.env.SUPERTONE_API_KEY });

const response = await client.textToSpeech.streamSpeech({
  voiceId: VOICE_ID,
  apiConvertTextToSpeechUsingCharacterRequest: {
    text: "This response is streamed chunk by chunk.",
    language: "en",
    model: "sona_speech_1",
  },
});

if (response.result && typeof response.result === "object" && "getReader" in response.result) {
  const reader = (response.result as ReadableStream<Uint8Array>).getReader();
  const chunks: Uint8Array[] = [];
  while (true) {
    const { done, value } = await reader.read();
    if (done) break;
    if (value) chunks.push(value);
  }
  fs.writeFileSync("streamed.wav", Buffer.concat(chunks));
}
Streaming is currently supported on sona_speech_1 only.

​
Long text auto-chunking

createSpeech and streamSpeech both auto-chunk text longer than 300 characters. Pass the text as-is — the SDK splits it, generates each segment, and merges (or streams) the result. 
const response = await client.textToSpeech.createSpeech(
  {
    voiceId: VOICE_ID,
    apiConvertTextToSpeechUsingCharacterRequest: { text: longText, language: "en" },
  },
  {
    maxTextLength: 300, // optional override (default 300)
  },
);
predictDuration does not auto-chunk — its 300-character limit is enforced. See Long text for details.

​
Common operations

// List voices with pagination
const list = await client.voices.listVoices({ pageSize: 20 });

// Search voices
const search = await client.voices.searchVoices({
  language: "ko,en",
  style: "happy",
});

// Get a single voice
const voice = await client.voices.getVoice({ voiceId: VOICE_ID });

// Predict duration (no credits deducted)
const { duration } = await client.textToSpeech.predictDuration({
  voiceId: VOICE_ID,
  predictTTSDurationRequest: {
    text: "How long will this be?",
    language: "en",
  },
});

// Get credit balance
const balance = await client.usage.getCreditBalance();

​
Type-safe enums (optional)

For type safety, the SDK exposes enum constants in @supertone/supertone/models: 
import * as models from "@supertone/supertone/models";

models.APIConvertTextToSpeechUsingCharacterRequestLanguage.En;     // "en"
models.APIConvertTextToSpeechUsingCharacterRequestModel.SonaSpeech1; // "sona_speech_1"
Plain string literals ("en", "sona_speech_1") work too — use whichever style you prefer.

​
Error handling

Errors live in @supertone/supertone/models/errors and all extend SupertoneError: 
import { Supertone } from "@supertone/supertone";
import * as errors from "@supertone/supertone/models/errors";

const client = new Supertone({ apiKey: process.env.SUPERTONE_API_KEY });

try {
  const response = await client.textToSpeech.createSpeech({ /* ... */ });
} catch (err) {
  if (err instanceof errors.TooManyRequestsErrorResponse) {
    console.log("Rate limited:", err.message);
  } else if (err instanceof errors.UnauthorizedErrorResponse) {
    console.log("Auth failed:", err.message);
  } else if (err instanceof errors.PaymentRequiredErrorResponse) {
    console.log("Out of credits:", err.message);
  } else if (err instanceof errors.SupertoneError) {
    console.log(`HTTP ${err.statusCode}: ${err.message}`);
  } else {
    throw err;
  }
}
Error classHTTP status
BadRequestErrorResponse400
UnauthorizedErrorResponse401
PaymentRequiredErrorResponse402
ForbiddenErrorResponse403
NotFoundErrorResponse404
RequestTimeoutErrorResponse408
PayloadTooLargeErrorResponse413
UnsupportedMediaTypeErrorResponse415
TooManyRequestsErrorResponse429
InternalServerErrorResponse500
Network-layer errors (ConnectionError, RequestTimeoutError, RequestAbortedError, etc.) extend HTTPClientError, not SupertoneError.

​
Configuration

const client = new Supertone({
  apiKey: process.env.SUPERTONE_API_KEY,
  timeoutMs: 30_000,
  retryConfig: {
    strategy: "backoff",
    backoff: { initialInterval: 500, maxInterval: 60_000, exponent: 1.5, maxElapsedTime: 3_600_000 },
    retryConnectionErrors: true,
  },
});

Python SDK

The equivalent SDK for Python.

Examples

Recipes for common workflows.