How Do You Add Sound Effects to an AI Video After It Is Made?

You add sound effects to a generated video with the Add Sound button that appears on a silent clip in your creations. Open the clip, describe the sounds you want, and the tool listens to what happens on screen and generates matching effects. The clip must be 30 seconds or shorter, and the price is 0.8 credits per second, shown before you confirm.

Updated October 2026.

Native audio versus adding sound afterwards

Some video models generate audio with the picture: footsteps when a person walks, an engine when a car moves, ambience in the background. When it works, that native track is the fastest route to a video that feels finished, as the generators with sound comparison explains. When it does not, you get audio that is generic, out of sync or simply wrong for the scene.

Adding sound afterwards gives you control. Instead of accepting whatever the model produced, you describe the sounds the scene needs and place them where they belong. That matters most for short social clips, where a well-placed impact or ambience does a lot of work in a few seconds.

The trade-off is time. A generated audio track arrives with the clip; added effects need one more step. For a batch of clips, that step is worth it when the sound is part of the point rather than background.

Add Sound step by step

Add Sound is a button on a video you generated in VdoBloom, not a separate dashboard tab and not an upload tool. It appears on eligible clips in My Creations.

1. Open a silent generated video. The button only works on a clip that has no audio track, or whose audio is effectively silent. A clip that already has sound is refused, because the tool would be adding a second effects layer over an existing mix.
2. Check the length. The clip must be 30 seconds or shorter. A longer clip is refused before any charge.
3. Describe the sounds. Write what should be heard, in plain language: the footsteps, the door, the room tone. The tool analyses the picture and places the generated effects against what happens on screen.
4. Confirm the price. The button shows the credit cost before you run it, based on 0.8 credits per second of the clip.
5. Generate and listen. The effect track is rendered onto the clip. A failed job is refunded.

Because the tool listens to the video rather than to a timeline you build, the description should name sounds that can actually be seen or implied by the action. “Wind” over a still indoor shot will not have much to attach to; “a kettle clicking off” over a close-up of a kettle will.

Writing foley prompts

Foley is the craft of recording the small sounds a scene needs. A prompt works the same way: name the source, the material and the intensity.

• Footsteps: “slow footsteps on a wooden floor, a person walking left to right”. Name the surface, because wood, gravel and carpet sound different.
• Ambience: “quiet cafe room tone, low chatter, a cup set down once”. Ambience should sit under everything else, so ask for something quiet.
• Impacts: “a heavy box landing on concrete, one thud, short tail”. Say how many and how long, or the result may repeat.
• Actions: “a zipper closing, fabric moving, close perspective”. Close perspective tells the tool the sound should feel near the microphone.
• Nature: “light rain on a window, steady, no thunder”. Add the negative when a common addition would ruin the scene.

Keep the description to one or two sentences. A long list of sounds gives the model more to place and more chances to put one where you did not want it. If a clip has three distinct actions, generate the dominant one first and judge the result.

These are prompts to try, not outputs. For a broader set of ideas, see our guide to AI sound effects and the best sound effect generators round-up.

Add a music bed on top

Effects make a clip feel real; music makes it feel like a piece. Once the effects are on the clip, open the Timeline Editor, add the clip to the video track, and put a music track on one of the two audio tracks. Each audio item has its own volume, fade in and fade out, so the music can sit under the effects rather than over them.

Generate the music in the music tab if you do not have a track. MiniMax Music 2.6 is 12 credits per song, and ACE-Step is 6 credits per generation. The full workflow, from mood to mix, is in adding music to a video with AI.

When effects and music share the same moment, lower the music. The effect is the detail the viewer is meant to notice; the music is the bed underneath it.

Put a voice-over over it

A voice-over is the third layer. Put it on the second audio track, or split the audio and place it where the narration runs. If the effects track is busy, the voice needs a clear space: lower the effects under the spoken section rather than raising the voice, and keep the music lower still.

Generate the voice in text to speech if you need one, then add it in the editor. The AI voice-over guide covers script length and pacing. A useful order is effects first, then voice, then music, because each layer is easier to judge against the one below it.

If the original clip already has native audio you want to keep, Add Sound will refuse it. In that case, take the clip into the Timeline Editor and build the mix there, where you can add music and voice on separate tracks without replacing the existing sound.

Sync and clipping mistakes, and clips over 30 seconds

Two problems account for most disappointing results.

• Sync. An effect that arrives a fraction late reads as detached from the action. Watch the clip with the picture in view and check the moment of contact: the foot meeting the floor, the lid meeting the box. If the effect is consistently late, the clip may need a trim rather than a new generation.
• Clipping. Layering effects, music and voice raises the overall level. If the loudest moment distorts, lower the music and effects rather than raising the voice, and leave headroom for the peaks.

Clips longer than 30 seconds cannot use the button. Split the clip into shorter sections, or move the whole job into the Timeline Editor, where up to two audio tracks and 180 seconds of output give you room to place effects and music by hand.

Add Sound is billed at 0.8 credits per second, with a floor of 1 credit for the job. A 10-second clip is 8 credits, and a 30-second clip is 24. The exact cost is shown on the button before you run it, and a failed job is refunded. The sound effects tab is the place to generate an effect you want to place yourself. Credit packs are one-time purchases that never expire, so a pack bought for one project is still there for the next; prices are on the pricing page.

Frequently asked questions

Can I upload my own video to get sound effects?

No. Add Sound is a button on a video you generated in VdoBloom. For your own footage, generate effects in the sound effects tab and place them in the Timeline Editor.

How long can the clip be?

30 seconds or shorter. Longer clips are refused before any charge; use the Timeline Editor for those.

Why is the button not available on my video?

It only applies to a silent clip. A video that already has audio is refused, because the tool would be layering effects over an existing mix.

How much does it cost?

0.8 credits per second, with a 1-credit floor. A 10-second clip is 8 credits and a 30-second clip is 24, shown before you confirm.

Can I add music and a voice-over as well?

Yes. Put the clip in the Timeline Editor and add music and voice on the two audio tracks, with volume and fades for each.

Open My Creations and choose a silent generated clip of 30 seconds or less, then describe the sounds you want in the Add Sound box. New accounts start with 10 free credits and no card required, enough for a short clip, so you can hear how a generated effect sits against your own picture before adding more layers in the Timeline Editor.