Ideas sound better out loud.

Give your ideas
a voice.

A workspace for speech, stories, and conversations.
Make something worth listening to.

Fine cyan glass strands woven into a flowing, sculptural sound wave
A little less typing. A little more feeling.

Hear an idea come to life.

Explore the voice library
Text to Speech
72/80
AI-generated audio · No account needed · Up to 80 characters. Request and daily limits apply. Replay uses audio cached in this page.Your words, ready to be heard.

One workspace. Every side of voice.

Create, listen, and build on your ideas.

A home for your next spoken idea.

Choose a voice, work on your words, and keep your creations together.

The real Helix Voice console in local demo mode, with a text editor, voice selector, and speech settings
Helix Voice console · Demo workspaceOpen console

A voice for what you’re building.

From the first word to the next conversation.

Let the story do the talking.

Find the voice for your next chapter, video, or audio guide. Shape your script and explore how it sounds before moving into production.

AudiobooksVideo voiceoversAudio guides
Explore Text to Speech
Narration concept
“Some stories start with a place. Others start with a voice you remember.”
Find a voice that fits your idea
For the builders

Your product.
With a voice.

Try the public short-text preview API and receive WAV audio. No account or API key is needed; request and daily limits apply.

Read the integration guide
Speech generation requests Audio transcription workflows One account across Helix Voice
POST/v1/audio/preview
// Anonymous short-text preview. Returns WAV audio.
const response = await fetch(
  "https://tts.picperweek.com/v1/audio/preview",
  {
    method: "POST",
    credentials: "omit",
    headers: { "Content-Type": "application/json" },
    body: JSON.stringify({
      input: "Every idea deserves to be heard.",
      voice: "demo"
    })
  }
);
if (response.status === 429) {
  // Wait for Retry-After before making another request.
  throw new Error("Preview limit reached: " + response.headers.get("Retry-After"));
}
if (!response.ok) throw new Error("Preview unavailable");
const url = URL.createObjectURL(await response.blob());
const audio = new Audio(url);
// This example plays once; the website caches audio for replay.
audio.onended = () => URL.revokeObjectURL(url);
await audio.play();
Anonymous preview · WAV audio · Respect Retry-After on HTTP 429.

A few things
you might be wondering.

A good place to start.
Then, let your ideas do the talking.

What can I do with Helix Voice?

Explore a workspace for text-to-speech, audio transcription, voice libraries, and voice agent configuration. You can also organize authorized recordings as voice cloning drafts. Model generation, transcription, cloning, and deployed agents depend on a connected backend service.

Can I try a voice without an account?

Yes. Enter up to 80 characters and choose a public preview voice to generate real AI audio. No account is needed. Request limits and daily allowances apply; the playground shows when to try again. Replaying audio already cached in this page does not send another request.

Which languages can I preview?

Try the English and Mandarin Chinese sample scripts, or enter your own short text. The preview service detects the language. Audio is generated using the selected public voice; pronunciation and language support depend on the speech model.

How do I access the console?

Choose Get started to create a Helix Voice account, or Log in if you already have one. Passport handles your account and returns you to the Voice console. The website, account pages, and console share the same Helix Voice identity.

Can I integrate speech into my own application?

The example on this page calls the public preview endpoint and receives WAV audio. It needs no login and is intended for short, limited previews. Use the console integration guide for authenticated speech generation and transcription workflows.

Your next idea
deserves to be heard.

Let’s give it a voice.

Get started