Background Audio
Background Audio is the ambient sound layer that gives scenes a sense of place — the hum of a busy street, rain on a window, the low noise of an empty hallway, a crowd in a bar. It plays beneath dialogue and music and is controlled independently from both.
ACT3 AI supports background audio as a separate rendering pass, giving you the option to isolate it from the visual render and control it with precision.
Background Audio as a Separate Render
For scenes that require clean audio post-production control, ACT3 AI can render a dedicated background audio pass — a second rendering that produces only the ambient and environmental sounds, without dialogue.
This separate pass is useful for:
- Post-production mixing in Premiere Pro, DaVinci Resolve, or Pro Tools
- Replacing or adjusting ambience without re-rendering dialogue
- Delivering separate audio stems for broadcast or theatrical delivery
- Layering your own custom ambient recordings over AI-generated dialogue
Enabling Background Audio Rendering
- Open Shot Settings for the shot (or Scene Settings for the whole scene)
- Go to the Audio tab
- Toggle Render Background Audio Track to ON
- Select the ambience type:
- Auto — ACT3 AI infers the appropriate background sound from the set description and scene context
- Manual — choose from the ambience library (see below)
- None — silence in the background audio pass (useful when adding your own ambience in post)
- When you render the shot, two audio files are produced:
- The main output (dialogue + ambience mixed together)
- A separate background audio stem (ambience only)
Both files are available in the Render Queue download after the job completes.
Background Audio Library
ACT3 AI includes a library of ambient sound categories you can manually assign to any shot:
| Category | Examples |
|---|---|
| Interior — Office | HVAC hum, keyboard sounds, distant phone |
| Interior — Home | Quiet room tone, ticking clock, faint traffic |
| Interior — Restaurant | Murmur of conversation, clinking glasses, soft music |
| Interior — Hospital | Distant beeping, intercom, quiet hallway |
| Exterior — Urban | Traffic, pedestrians, distant sirens |
| Exterior — Nature | Forest birds, wind through trees, stream |
| Exterior — Beach | Waves, seagulls, wind |
| Weather | Rain on windows, thunderstorm, heavy rain exterior |
| Vehicles | Car interior road noise, airplane cabin hum |
| Crowd | Large crowd murmur, applause, sports stadium |
Search the ambience library by keyword in the Audio → Ambience panel.
Turning Background Audio On or Off
Background audio is enabled by default for all shots. To turn it off:
- Open Shot Settings
- Go to the Audio tab
- Toggle Background Audio to OFF
Turning it off produces a shot with only dialogue — no ambient layer. This is appropriate for highly controlled studio scenes or if you intend to add your own ambience in post.
You can toggle background audio at the project level under Project Settings → Audio Defaults to set a default for all new shots.
Controlling Ambience Volume
In the editor's Audio panel:
- Adjust the Ambience Volume slider (0–100%) for each shot independently
- Use the Fade In and Fade Out controls to smoothly ramp ambience at the start or end of a shot
- The ambience volume is relative to the dialogue level — set it so ambience is present but does not compete with speech
For most interior dialogue scenes, 20–40% ambience relative to dialogue produces a natural result. Exterior scenes with no dialogue can run at 80–100%.
Best Practices
- Auto-assign ambience for first drafts — manual adjustment is only needed when the automatic selection does not match the scene
- Use the separate background audio stem whenever delivering to broadcast, streaming platforms, or professional post-production workflows
- Turn off background audio for dream sequences, memory sequences, or any stylized moment where silence is part of the design
- Keep interior background audio subtle — overpowering ambience distracts from dialogue and makes the scene feel artificial
Related
- Video Editor — parent
- Introduction to ACT3 AI — grandparent
- TTS and Audio
- Voice Casting
- Render Queue
- Export