Skip to main content

Browse guides

Open the full guide library for quick switching.
Audio & Voice 6 min read Updated April 10, 2026

Audio & Text-to-Speech: Voices, Formats, and Delivery

Run the audio workflow with one exact script, one exact voice, and a matching MP3 export example.

Audio mode with the selected voice and MP3 format visible

Audio mode is the fastest way to show that the widget can generate something useful beyond visuals. The best demos are short, concrete, and spoken the way a real listener would actually hear them.

What you’ll learn

  • Which exact voice and format were used in this walkthrough
  • What script is shown in the prompt field for this walkthrough
  • How to explain Voice and Format without overcomplicating them
  • Which exact result actions appear after generation finishes

Before you start

  • Open Generate
  • Switch to Audio mode
  • Keep the first demo on MP3 with one short narration script
01 Set the output

Set Voice to Kore - Firm (F) and keep Format on MP3

Click Audio in the mode selector, then match the same two settings shown in the screenshot.

  • Under Voice, choose Kore - Firm.
  • Under Format, keep the first demo on MP3.
  • Explain that Voice changes delivery style, while Format changes the exported file type.
  • Only switch to WAV if the customer has a specific production reason for it.

Expected result

The customer knows exactly what kind of spoken delivery they are about to generate before the script is even pasted.

Audio mode with Kore - Firm selected and MP3 chosen
The empty state now shows prompt suggestions above the audio controls, but the core setup is still one voice and one export format the customer already recognizes.
02 Write the script

Paste one short narration script that sounds like real speech

Paste this script into the textarea:

See every KPI move in real time with Advanced Image Analytics. Turn campaign data into a premium product story in seconds.
  • Write in full sentences, not fragmented notes.
  • Keep the first script to one or two short lines.
  • Use punctuation normally so the generated delivery has natural pauses.
  • Keep the model shown in the screenshot selected for this walkthrough.
  • Click Generate once the script reads like real speech.

Expected result

The generation request is specific enough that the final audio can be judged immediately instead of sounding like a generic placeholder.

Audio mode showing a short narration script pasted into the textarea
The input should already sound like spoken language. That makes the final voice easier to judge in one listen.
03 Review the result

Review the player, then keep or revise the read

Once the player appears, review the output exactly the way the customer would consume it: listen all the way through, then choose the next action from the footer.

  • Use Download when the read is already usable.
  • Use Generate again when the script is fine but the delivery still feels off.
  • Use Edit prompt if the wording itself needs work.
  • If tone is the problem, change Voice before regenerating.

Expected result

The team can decide whether the spoken line is already usable or whether the script or voice choice needs one more pass.

Sample generated audio

Listen to the sample output

This guide includes the exported MP3 example used alongside the walkthrough so the reader can hear the same voice, format, and delivery style referenced above.

The finished audio result player ready for playback and download
The review step is about the read itself: pacing, tone, and whether the spoken message sounds launch-ready.
  • One default voice that feels broadly usable, such as Kore - Firm
  • MP3 as the first export format
  • One short reusable script for demos, screenshots, and onboarding
  • Video Generation if your team uses narration on generated clips
  • Text Generation if you also want to show script creation inside the same widget
  • Docs for audio-related widget events and configuration

Previous guide

Video Generation: Clips, Queueing, and Voiceovers

Run the live video flow from mode selection and motion prompt to in-progress preview and the optional voiceover pass, then compare it with the final exported clip.

Next guide

Branding & Guidelines: Control the Style Before and After Generation

See how brand guidelines influence generation and how the branding editor finishes assets with overlays, logos, and a final branded export.