How audio works in animated scenes

When you animate a scene using a model that includes audio, your video can have two sources of sound:

1. Narration (voiceover)
   → Generated from your script

2. Background audio (from the animation model)
   → This may include ambient sound, music, or even spoken dialogue

Not all animation models include audio

Some animation models generate background audio, while others generate visuals only.

• Models with audio can include ambient sound, music, or dialogue

• Models without audio will rely entirely on your narration

If you’re hearing overlapping audio, you’re likely using a model that includes background audio.

Why you might hear double or overlapping audio

If both are active at the same time, you may notice:

• The same words spoken twice

• Dialogue layered on top of narration

• Audio that feels cluttered or unusable

This happens because your script narration and the model’s background audio are both playing simultaneously.

How to fix it

You can control audio per scene using the … (more actions) menu in the timeline.

Option 1: Mute narration

• Select Mute scene

• This removes the voiceover for that scene

• Keeps the animation’s background audio (including dialogue)

Best when you want the animation to drive the audio

Option 2: Mute background audio

• Select Mute background audio

• Keeps your script narration

• Removes audio generated by the animation model

Best when you want clean, controlled voiceover

Using animation audio for dialogue

Some animation models can generate spoken dialogue as part of the background audio.

If you want to use this:

• Make sure each scene contains a complete line of dialogue

• Avoid splitting one sentence across multiple scenes

This helps the model generate more natural, complete speech. If dialogue is split across scenes, the generated audio may feel:

• choppy

• cut off

• or unnatural

Keeping each scene as a full thought leads to better results.

Works with both animation and Full Motion

This behavior applies to:

• scene-level animation (image → video)

• Full Motion videos

In both cases, audio can come from:

• your script (narration)

• the model (background audio)

Summary

• Animated scenes can include two audio sources

• Some models include background audio, others don’t

• Overlapping audio happens when both are active

• Use Mute scene or Mute background audio to control it

• For best results, keep dialogue contained within each scene