SDKs & Libraries

Use our official SDKs to integrate VoiceMaker into your applications with minimal code. Both libraries are published as myvoicemaker and cover all five modules: ASR, TTS, Animate, Explain, and Usage.

Python SDK

Installation

cURL

pip install myvoicemaker

Quick start

Python

from myvoicemaker import VoiceMaker

client = VoiceMaker(api_key="vmk_live_a1b2c3d4e5f6_8h9i0jKLmnOPqrSt")

# Generate speech
result = client.tts.generate(
    text="Ẹ kú àárọ̀.",
    voice_id="amara-okafor-female-conversational",
    language="yo",
)
print(result.audio_url)

# Transcribe and wait for the result
job = client.asr.transcribe("https://example.com/audio.wav", language="ha")
final = client.asr.poll(job.job_id, timeout_seconds=120)
print(final.text)

Available methods

Method	Description
`client.asr.transcribe(...)`	Submit a transcription from a URL
`client.asr.transcribe_file(...)`	Submit a transcription from a local file
`client.asr.get_result(job_id)`	Fetch a transcription result
`client.asr.list(...)`	List transcription jobs
`client.asr.poll(job_id)`	Wait for a job to complete
`client.asr.stream(...)`	Real-time streaming transcription over WebSocket
`client.tts.list_voices(...)`	List available voices
`client.tts.generate(...)`	Generate speech (`mode="async"` queues the job instead)
`client.tts.get_result(id)` / `client.tts.poll(id)`	Fetch / wait for an async TTS job
`client.tts.stream(...)`	Real-time streaming speech synthesis over WebSocket
`myvoicemaker.chunk_text(...)`	Split text into sentence-sized chunks for streaming
`client.animate.generate(...)`	Submit a lip-sync animation
`client.animate.get_result(job_id)`	Fetch an animation result
`client.animate.poll(job_id)`	Wait for an animation to complete
`client.usage.get_balance()`	Get the credit balance
`client.usage.get_breakdown(...)`	Get usage by module
`client.explain.process(...)`	Explain, summarize, translate, or simplify text

Node.js SDK

Installation

cURL

npm install myvoicemaker
# or
bun add myvoicemaker

Quick start

JavaScript

import { VoiceMaker } from "myvoicemaker";

const client = new VoiceMaker({ apiKey: "vmk_live_a1b2c3d4e5f6_8h9i0jKLmnOPqrSt" });

// Generate speech
const result = await client.tts.generate({
  text: "Ẹ kú àárọ̀.",
  voice_id: "amara-okafor-female-conversational",
  language: "yo",
});
console.log(result.audio_url);

// Transcribe and wait for the result
const job = await client.asr.transcribe({
  audio: "https://example.com/audio.wav",
  language: "ha",
});
const final = await client.asr.poll(job.job_id, { timeoutMs: 120_000 });
console.log(final.text);

// Real-time streaming (browsers, Node >= 22, Bun, Deno)
const stream = client.asr.stream({ language: "yo" });
stream.on("transcript", (t) => console.log(t.is_final ? "FINAL" : "…", t.text));
stream.sendAudio(pcmChunk); // PCM16LE mono
await stream.close();

// Real-time streaming TTS: hear audio while later text still synthesizes
import { createPcmPlayer } from "myvoicemaker";
const tts = client.tts.stream({ voiceId: "amaka", language: "auto" });
const player = createPcmPlayer(); // browser Web Audio helper
tts.on("audio", ({ data }) => player.enqueue(data));
tts.sendText("Barka da zuwa. Yaya kake a yau?");
await tts.close();

The JavaScript SDK exposes the same modules with camelCase method names (listVoices, getResult, getBalance, getBreakdown) and typed error classes (AuthenticationError, InsufficientCreditsError, RateLimitError, and more). Errors in both SDKs name the SDK method that failed and carry the API's errorCode as err.code / e.code — see Error Codes for the catalog and the class each code maps to. Streaming helpers chunkText, createPcmPlayer (browser playback) and pcmChunksToWav (WAV assembly) ship with the package. See Transcription → Real-time streaming and Text-to-Speech → Real-time streaming for the full streaming protocols.

REST API only

If your language isn't listed above, you can call the REST API directly — every endpoint is documented in this reference. Both SDKs default to the base URL https://api.myvoicemaker.ai and target the /dev/v1 routes.