Nodaro DocsMix Audio
Layer several audio tracks into one file, with a volume from 0 to 200% for each, and let music dip automatically under a voiceover while it speaks.
The Mix Audio node layers several audio tracks into one file, with a volume for each track from 0 to 200%. Use it to put a voiceover over background music, or sound effects over an ambience, and then send the mix to Merge Video & Audio.
With Duck the others under set, one track, usually the voice, pulls every other track down while it is loud. The other tracks come back up in its pauses. A music bed then sits under the speech, and you do not have to adjust its volume by hand.
When to use it
- Put a voiceover on top of background music.
- Combine sound effects with a music track.
- Mix dialogue with ambience for a cinematic scene.
- Build an audio bed from several generated tracks.
Quick start
Add the node
Press Tab on the canvas and choose Audio › Edit Audio › Mix Audio.
Connect the tracks
Wire two or more audio nodes into the Audio input.
Set the volumes
Open the settings panel. Each connected track has its own volume slider, from 0 to 200%. With more than one track, you can also drag the tracks into order.
Run it
Click Run on the node. The mix appears on the node with a player.
Input and output
| Input | Accepts | What it does |
|---|---|---|
| Audio | Two or more audio or video nodes | The tracks to layer. Each connected node appears in the track list. A connected video counts as a track: its sound is mixed, and a video with no sound adds silence for its length. |
The output, Audio, is one mixed file.
Settings
| Setting | What it does |
|---|---|
| Track order | The connected tracks, listed by node name. Drag to reorder them. |
| Volume, per track | From 0 to 200%. The default is 100%. |
| Duck the others under | The track that the other tracks dip under. The default is Off, a plain mix. The setting appears when two or more tracks are connected. |
| Duck amount | How far the other tracks dip, from 0 to 100. The default is 75. The setting appears when a duck track is set. |
| End the mix | Where the mixed file ends: With the longest track (the default) or When the voice ends, which stops the mix when the Duck the others under track ends. The setting appears when a duck track is set. Turning the duck off resets it. |
Duck music under a voice
Ducking lowers the other tracks while one track is loud. The loudness of the Duck the others under track controls a compressor on every other track. While that track is louder than -30 dBFS, the other tracks are pulled down. When it falls quiet, they return to their own level.
Duck amount sets how much of the voice's level above that threshold is taken off the other tracks. A higher value gives a deeper dip:
| Duck amount | Compressor ratio |
|---|---|
| 0 | 1, no dip |
| 50 | 2 |
| 75, the default | 4 |
| 90 | 10 |
| 95 to 100 | 20, the maximum |
The ratio is 1 / (1 − amount / 100), with a maximum of 20. With the default of 75, the ratio is 4. The music starts to dip within about 20 ms of the voice, and it comes back over about 500 ms after the voice pauses.
- The voice keeps its level. A plain mix divides every track by the number of tracks. A ducked mix adds the tracks together instead, and a limiter keeps the sum from clipping.
- The music plays to its end, unless you end the mix with the voice. When the duck track ends before the others, the other tracks return to full level and play to their own end. Set End the mix to When the voice ends to stop the whole mix when the duck track ends, so a long music bed ends with the episode.
- The price stays the same. A run costs the same with or without a duck.
If the duck track is disconnected later, the node makes a plain mix.
Tips
- Keep speech clear. Leave the voiceover at 100% and set background music between 30% and 50%. Or set Duck the others under to the voiceover, so that the music dips only while the voice speaks.
- Duck under the voice, not the music. Choose the voice track in Duck the others under. Raise Duck amount if the music still competes with the speech.
- Boost with care. A volume above 100% can clip and distort. Level quiet sources with Adjust Volume first.
- Layer on purpose. Start with the voice, then add the music, then the effects.
- Listen before you move on. Check the mix on the node before you connect it to the next node.
- One track only? To change the volume of a single track, use Adjust Volume instead.
From the API
POST /v1/mix-audio takes audioUrls, a list of 2 to 20 tracks, and the optional trackVolumes, duck and endAt. endAt is "longest" (the default) or "voice", which ends the mix when the duck.under track ends. "voice" needs duck; without it the request is refused with 400 validation_error. duck takes under, the 0-based index of the track that the others dip under, and the optional amount (default 75). It also takes these fine controls, which the node always leaves at their defaults:
| Field | Default | Range |
|---|---|---|
thresholdDb | -30 dBFS | -60 to 0 |
attackMs | 20 | 1 to 2000 |
releaseMs | 500 | 10 to 9000 |
ratio | From amount | 1 to 20. Overrides amount when you set it. |
The MCP tool is mix_audio, where the same setting is end_at. See Voice and media.
Frequently asked questions
How loud should background music be under a voiceover?
Keep the voiceover at 100% and set the music between 30% and 50%. The speech then stays clear over the music.
Can I make a track louder than the original?
Yes, up to 200%. Use values above 100% with care, because they can clip and distort the sound.
What is the difference between Mix Audio and Combine Audio?
Mix Audio plays the tracks at the same time, layered on top of each other. Combine Audio plays clips one after another.
How many tracks can I mix?
Connect two or more audio nodes. Through the API, one mix takes from 2 to 20 tracks.
Can the music dip automatically while the voiceover speaks?
Yes. Set Duck the others under to the voiceover track. Every other track is pulled down while the voice is loud and comes back up in its pauses. Duck amount, 75 by default, sets how far the music dips.
Related
Combine Audio
Adjust Volume
Merge Video & Audio
Text to Audio
Last updated on
Combine Audio
Join audio clips end to end into one file, in the order you choose, and trim each clip first. For narration, podcasts and dialogue made in several parts.
Adjust Volume
Change the volume of an audio track or of a video's sound, level it with Normalize, and add a fade-in and a fade-out of up to 10 seconds each.