Skip to content

PyAI Cast / Expressive voice production

A good script.
A great performance.

A little warmth. A moment of suspense. A line that lands. Auto-Director gives every spoken line a direction, so your voiceover moves with the story.

One stock or cloned voice for the whole production. Every line’s emotion and intensity stays editable.

The performance deskPress play on a line
The Lantern
3 spoken lines
Auto-Directed
1

The city had forgotten what silence sounded like.

deepbold
2

Until one small lantern appeared in the window.

warmbalanced
3

And then the whole skyline answered.

excitedbold
Voice: Ava Open Cast

Recorded Cast line previews. Listen to the delivery here, then open Cast to direct and render your own script.

Auto-Director for the first pass
Emotion and intensity, line by line
Preview, render and download

From the page to the performance

The direction comes
before the recording.

Give the script a complete first pass. Keep the moments you like, change the ones you don’t, and hear a line before committing to the full production.

01

Bring the words

Paste a scene, episode, lesson or full narration script.

02

Find the direction

Auto-Director assigns emotion and intensity to each spoken line.

03

Hear the moment

Preview a line and adjust the delivery where it matters.

04

Keep the performance

Render asynchronously, follow the job and download the finished WAV.

Same voice. A different feeling.

Let the delivery
follow the writing.

A reassuring line and a cliffhanger need different performances. Give each moment its own emotion and intensity while keeping one narrator across the whole piece.

Explore a few direction plans. These examples show how to approach a script; the recorded line previews above let you hear Cast.

Back to the listening desk
A script, with direction

An emotional turn, with one narrator throughout.

01

“You made it.”

warmbold
02

“I was beginning to think the storm had won.”

empathysubtle

Illustrative direction plans. Preview recorded performances above, or render your own script in Cast.

Made for a story with an arc

Direct the moments
people remember.

From the opening sentence to the final sign-off, build a performance that gives your audience a reason to keep listening.

01

Storytelling

Give narration an emotional arc without directing every sentence by hand.

02

Podcasts

Move from a strong cold open to a thoughtful explanation without one flat read.

03

Audiobooks

Direct scene changes and dialogue beats across chapters while keeping one narrator.

04

Product videos

Keep feature explanations clear, then lift the delivery when the payoff arrives.

05

Training and learning

Make instructions easy to follow and sensitive moments sound considered.

06

Brand audio

Keep campaign reads recognizable while changing the energy line by line.

Built for your production pipeline

Choose a delivery.
Take the code.

Change the words, emotion and intensity. Your request updates as you direct the line. Copy it to your server, or continue in the full Cast workspace.

Direct a line. Copy the request.

Choose the words, voice and delivery for a Cast preview. The examples save a WAV on your server.

One short line. Use render jobs for long-form production.

POST /v1/cast/speech

import { writeFile } from "node:fs/promises";

const response = await fetch("https://api.pyai.com/v1/cast/speech", {
  method: "POST",
  headers: {
    Authorization: `Bearer ${process.env.PYAI_API_KEY}`,
    "Content-Type": "application/json",
  },
  body: JSON.stringify({
    "text": "Welcome to The Signal. Today, a small idea opens up a whole new world.",
    "voice": "stock_ava_en_us",
    "emotion": "warm",
    "intensity": "balanced"
  }),
});
if (!response.ok) throw new Error(await response.text());
await writeFile("cast-preview.wav", Buffer.from(await response.arrayBuffer()));

Code updates here without sending a request. Run the copied example on your server with PYAI_API_KEY.

Copy the code to use these settings, or explore the full product workspace.

The builder generates code without rendering audio. The Cast API also supports directing a script and tracking long-form render jobs. Read the Cast guide ↗

A little more detail

Your next
production, explained.

Cast is $0.02/min of finished audio, measured per second. See current account terms before running a production render.

Explore pricing and terms

Need immediate synthesis for a single prompt or conversational response?

Explore Speak
What does Auto-Director do?

Auto-Director reads a multi-line script and assigns an emotion and intensity to each spoken line. You can keep the plan, change any direction, or preview one line before rendering the full piece.

Can I use my own voice?

Yes. Cast uses the same PyAI voice catalog and cloning tools as Speak. Choose a stock voice or a voice you have cloned, then use it across the whole production.

How does long-form rendering work?

A render runs as an asynchronous job. Cast shows progress, saves the job to your project, and gives you a WAV player and download when the audio is ready. You can close the tab and return later without losing the render.

How much does Cast cost?

Cast is $0.02/min of finished audio. Usage is measured per second, so a short production is billed for its actual duration rather than rounded up to a full minute.

When should I use Cast instead of Speak?

Use Cast when delivery changes across a script and you need to direct emotion line by line. Use Speak when you need low-latency synthesis for a single prompt or a conversational response.

Can developers use the Cast API directly?

Yes. The same workflow is available through the Cast API: direct the script, preview a line, submit an asynchronous render job, poll its progress, then fetch the finished audio.

The words are yours. So is the direction.

Give your next story
something to say.

Preview a line, shape the performance and render the finished voiceover from your PyAI account.