OneTaskBy Tim King · Updated
Audio Ducking
Auto Duck Music Under a Voiceover or Podcast, Then Export the Mix
OneTask is free, with no sign-up, and built by one person.
If this tool saved you time, a coffee helps keep new ones coming.
How the Audio Ducking works
Audio ducking lowers background music automatically while someone is speaking and brings it back up in the gaps. Add the voice file first and the music second, choose how far and how fast the music dips, and export a finished mix for a podcast, video or voice-over.
How audio ducking works
The tool measures the loudness of the voice track every ten milliseconds. When it crosses the threshold, the music is pulled down by the duck amount; when the voice stays quiet for 300 ms the music rises again. Attack and release smooth those moves so they sound like a person riding a fader, not a switch. The voice and the ducked music are then mixed into one file.
Audio ducking settings
Set the threshold just above the background noise of the voice recording so breaths do not trigger it; minus 40 dBFS suits most clean recordings. Twelve to eighteen decibels of duck keeps music audible but clearly behind speech. A short attack of 100 to 200 ms gets out of the way fast; a longer release of half a second or more avoids the music pumping between words.
File order
Use the up and down buttons so the voice is first and the music second. The mix is as long as the longer file, so trim the music beforehand if it runs past the end of the talking.
FAQ
Can I use more than two files?
Only the first two are used: file one is the voice that controls the ducking and file two is the music that gets ducked. Mix several voices together first with the audio mixer if you need more.
Why does the music pump up and down between words?
The release is too short, so the music rises in every small pause. Increase the release to around 800 ms or more, which keeps the music down through natural gaps and lifts it only in longer breaks.
Will it change the voice volume?
No. The voice passes through unchanged and only the music is turned down. Use the music level slider to set the overall balance, and normalise the voice beforehand if it is too quiet.
See also: all Audio tools, Images, Video & Screen, PDF, SVG, Simulations & Screensavers, Generative Art, Chart Makers, 3D Model Viewers, Drawing Pads, Mockups & 3D Making, Diagram Makers, 3D Science, Tracker Music.
Related tools
Audio Mixer
Layer several audio files to play together, with clip-safe levels.
Audio Crossfader
Blend tracks together with smooth, adjustable crossfades.
Audio Compressor (Dynamics)
Compress dynamics with threshold, ratio, attack and release.
Audio Normalizer
Normalise audio to a peak dBFS or LUFS loudness target, in browser.
8D Audio
Rotate any song around the listener's head, 8D style.