Give your ideas
a voice.
A workspace for speech, stories, and conversations.
Make something worth listening to.

Hear an idea come to life.
One workspace. Every side of voice.
Create, listen, and build on your ideas.
Words, brought to life.
Choose a voice and shape the pace of your next narration in the Text to Speech workspace.
Keep the words that matter.
Prepare audio and explore transcription with a connected speech service.
Let’s make something
worth listening to.
A voice of your own.
Organize authorized voice recordings and prepare a custom voice profile for your service.
Start a better conversation.
Give an agent a voice, a purpose, and your instructions. Preview a greeting in the console.
A home for your next spoken idea.
Choose a voice, work on your words, and keep your creations together.

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.
Explore Text to SpeechYour 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/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();
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