Skip to main content

AI video with sound. Make it feel there.

Use an AI video generator with sound to direct the picture, the action and the atmosphere in one brief.

Describe what to make with Video
≈ – credits
Try a prompt:
Beyond the Screen
Beyond the Screen

Create a new vertical 10-second cinematic screen-break illusion. Begin behind an audience in a dark cinema with red seats. A huge realistic dinosaur approaches through a misty forest on the cinema screen, then its head appears to cross the screen boundary into the auditorium. Nearby viewers duck and turn in surprise as the camera shifts toward a seated viewer, keeping the dinosaur close above the seats. Build tension, then a comic reaction; maintain coherent creature scale, theatre geometry and light. Choose a new dinosaur design and audience rather than copying the exact faces. Add a suitable original suspense-and-reaction soundtrack.

10 s

Cinematic sound effects and wordless audience reactions; no intelligible dialogue.

How it works

Try it now
  1. 01

    Say what you want

    A sentence or two is enough.

  2. 02

    Work with HowWorks Agent

    Describe the result you want. The Agent works with the tools you choose.

  3. Beyond the Screen
    03

    Make it

    Start creating, then keep working with HowWorks.

Explore the possibilities

Beyond the Screen
Cinematic sound effects and wordless audience reactions; no intelligible dialogue.
Gothic Hall Dance
Every single night, I pray that my [unclear] are sitting right. [unclear], my corset is so tight. Baby girl, [unclear], I feel like Nicole.
Upside-Down Garden Dance
Rhythmic dance music with repetitive wordless vocal sounds; no intelligible dialogue.
The Straight Line
He knew this door led downstairs. He opened it and died below. Footage shows him leave a safety marker, then curve towards the stairwell without visibly turning. Tracking was normal. The new build promised to steer him to safety. Why did it steer him here?
Mika // LIVE
Whoops! Total... Intentional drift! You guys love that, don't you? Ashen? Hey, it's me. Keep this line open, yeah?
Adrian Vale
You're late. I've been waiting a long time.
Unclaimed
They forgot him, like he was never even there. But I remember. We remember. And he's just still waiting. And he's still waiting.
Heirs
Move. 16 years gone, and I'm not just home. [Unclear phrase; the approved subtitle reads: “I’m the wild card.”] The game isn't about winning. It's about who gets to rearrange the chairs.
The President’s Nanny
I make beds, I remove stains, I'm just a contract worker. But one day, I picked up the wrong folder. Good, you're here. The state dinner seating chart needs your touch. Dad, she's not the aide. She changes my sheets. Now I'm solving state-level problems with a lint roller.
You Clicked Accept
19 years, the world has revolved around gates. But the real game is the fine print. Hey, what are you doing?

About

Create a video scene with audio that belongs to the action: rain against a window, a cup meeting a table or footsteps crossing a quiet room. Tell the HowWorks Agent what happens in frame, then describe the ambience, action sounds and any speech or music you want. This starting point uses Seedance 2.5 with sound enabled. Watch and listen to the result together before placing it in a finished video. Use the next message to refine the scene or ask for a different take.

Questions

How do I prompt an AI video generator with sound?

Describe the subject, one main action and the camera. Then add an audio direction: what makes a sound, when it happens and what stays in the background. Say no music or no speech when those distinctions matter.

Which model does this page start with?

The starting preset uses Seedance 2.5 with sound on. Review the selected model, available settings and estimate before sending. The sound and image still need to be checked in the delivered result.

Can I request dialogue or a specific sound effect?

You can include a short spoken line or describe an audible action. Listen for the actual wording, pronunciation and timing, and inspect any visible mouth movement. A written request does not guarantee exact speech or synchronization.

Can I use this to add sound to a video I already have?

This page generates a new video from a written scene. It is not an audio-dubbing or soundtrack-editing workflow for existing footage. Prepare separate sound work in your editing process when you need to preserve an existing clip.

What should I check before using the clip?

Watch once for the action, then listen for the sound and watch again for their relationship. Check the whole clip for unexpected speech, music or noise, and test it in the intended edit before publishing.