what is audio ducking

What is audio ducking and how to use it

Quick answer: Audio ducking is a technique that automatically lowers the volume of one audio track, usually music, whenever another track, usually a voice, is playing. The music dips under the speech and rises again in the pauses, so the voice stays clear without anyone having to ride the faders by hand. It is used in podcasts, videos, radio, broadcasts and music mixes, and most DAWs can do it with a sidechain compressor or a dedicated ducking tool.

If you have ever listened to a podcast where the intro music fades neatly under the host’s voice, or watched a video where the background track drops away the moment someone starts speaking, you have heard audio ducking. But what is audio ducking exactly, and how do you set it up yourself? In simple terms, it is automatic volume control driven by another sound. One track, called the trigger or key, tells a processor on a second track when to turn down. When the trigger stops, the second track returns to its normal level.

Ducking matters because speech and music compete for the same space in a mix. If the music stays at full volume under a voice, words get buried and listeners strain to follow. If you turn the music down permanently, the mix feels flat and lifeless. Ducking solves both problems: the music is present and full in the gaps, and it steps aside exactly when words need the room.

How does audio ducking work?

Ducking is usually built on sidechain compression. A compressor on the music track is told to listen not to the music itself, but to the voice track — the sidechain input. When the voice crosses a set threshold, the compressor reduces the music’s volume by an amount set by the ratio. When the voice stops, the compressor releases and the music returns to full volume.

Some DAWs and editors offer dedicated ducking tools that hide this complexity. Instead of setting up a sidechain yourself, you select the voice track and the music track, choose how much the music should dip, and the software handles the routing. How AI is changing audio editing describes how newer tools are making this kind of routine work faster, and automatic ducking is one of the clearest examples.

What is the difference between ducking and sidechain compression?

Ducking and sidechain compression are closely related, and the terms are often used interchangeably, but they are not quite the same thing. Sidechain compression is the technical method. Ducking is the result you want — one sound making way for another. You can achieve ducking with a sidechain compressor, with volume automation, or with a dedicated ducking plugin. Sidechain compression, meanwhile, is also used for creative effects that have nothing to do with speech, such as the pumping bass in electronic dance music, where the kick drum triggers a compressor on the bassline.

The practical difference is intent. In ducking, the goal is transparency: the listener should notice the voice, not the volume change. The dip is usually gentle, the attack is fast enough to protect the first syllable, and the release is slow enough to sound natural. In creative sidechaining, the pumping itself is part of the sound, so the settings are deliberately extreme. If your ducking is audible as a rhythmic pumping, you are closer to an effect than to a mix technique, and the release or the amount of reduction probably needs softening.

Volume automation is the manual alternative to both. Instead of a compressor reacting automatically, you draw the music’s volume down under each line of speech and back up in the gaps. Automation gives complete control and can sound extremely natural, but it is slow on a long podcast or video. Most creators use automatic ducking for the bulk of the work and then fix any awkward moments with a little automation afterwards. That combination gives both speed and precision.

Where is audio ducking used?

Podcasting is the most common home for ducking. A podcast usually has a voice track and at least one music track — an intro, an outro, and sometimes a bed under the conversation. Ducking keeps the intro music lively while making sure the host is always intelligible. It is especially useful when a show has multiple voices, because the music responds to whoever is speaking. If you are setting up a show, the how to start a podcast in 2026 guide covers the recording setup that ducking builds on.

Video production uses ducking in the same way. A YouTube video, a documentary, or a corporate film often has narration over music and ambient sound. The narration is the priority, so the music ducks under it and returns in the visual-only sections. Sound design for film uses the same idea more subtly: dialogue stays forward while effects and score sit underneath, as explored in how pet sounds and animal voices are used in sound design, where layered sounds must share space without masking the story.

Radio and broadcasting rely on ducking constantly. A DJ’s voice over a song intro, a newsreader over a theme, or a sports commentator over crowd noise all use the same principle. In live sound, ducking can keep announcements clear over background music in a venue. In music mixing, ducking is used more sparingly — for example, lowering a pad slightly when a vocal enters — because the goal is usually a balanced blend rather than a clear hierarchy. The tips to give your mix the Nugen Stereoizer treatment show how small mix moves add up, and ducking is one of those small moves.

How do you set up audio ducking in a DAW?

Setting up ducking in a DAW takes a few minutes once you know the routing. The steps below describe the standard sidechain method, which works in most DAWs, including Pro Tools, Logic Pro, Reaper and Audacity with a compressor that supports sidechaining.

First, put the voice and the music on separate tracks. Ducking cannot work if both sounds share a track, because the processor needs to tell them apart. Make sure the voice track is clean: trim silence, remove obvious mouth clicks if your editor supports it, and balance the level so it is consistent. A voice that jumps in volume will trigger the ducking unevenly.

Second, add a compressor to the music track and enable its sidechain input. Select the voice track as the sidechain source. In some DAWs this is a menu on the compressor; in others it is a send from the voice track to the compressor’s key input. Check your DAW’s documentation for the exact routing — the concept is identical everywhere, but the buttons differ. The sound engineering fundamentals course covers routing and signal flow in more detail if this step feels unfamiliar.

Third, set the threshold so the compressor reacts to normal speech but not to background noise or breaths. Play the voice track and watch the compressor’s gain reduction meter: it should move when someone is speaking and sit still in the gaps. Fourth, set the amount of reduction. For a podcast, a dip of around 6 to 10 dB usually keeps the voice clear while leaving the music audible. For a video bed, a smaller dip of 3 to 6 dB may be enough.

Fifth, set the attack fast, often under 10 milliseconds, so the first word is not masked. Set the release to a medium value, often 200 to 500 milliseconds, so the music returns smoothly rather than snapping back between words. Finally, listen to the whole section, not just a few seconds. Ducking problems usually show up over a full sentence or a full minute, not in a short test.

Which ducking settings matter most?

Threshold is the first setting to get right. Set it too high and the compressor never engages, so the music never dips. Set it too low and it reacts to breaths, room noise and quiet syllables, making the music wobble constantly. The right threshold sits just below the level of normal speech and above the level of the gaps between words. A compressor’s gain reduction meter is the best guide here: it should move with speech and rest in silence.

Ratio controls how much the music dips once the threshold is crossed. A low ratio gives a gentle dip that may not be enough for a loud music bed, while a high ratio gives a strong dip that can sound abrupt. For speech, a moderate ratio usually works well. If the music still masks the voice at a moderate ratio, lower the music’s overall level first rather than pushing the ratio higher — ducking works best when the mix is already roughly balanced.

Attack and release shape how natural the result sounds. A fast attack protects the start of each sentence. A release that is too fast causes pumping, where the music jumps up between words, while a release that is too slow leaves the music quiet after the voice ends, which sounds dull. If your DAW’s compressor has a lookahead control, a small amount can help the dip start just before the voice, which is useful for very fast speech.

Finally, consider the music itself. A dense, loud track with a strong beat needs more reduction than a sparse piano bed, and a track with lyrics will clash with speech far more than an instrumental, so choose instrumental beds for voiceover work. Ducking is one of the essential audio processing tools in a creator’s kit, but it is not a fix for a bad recording — if the voice track is noisy or uneven, fix that first.

SettingStarting point for speechIf it sounds wrong
ThresholdJust below normal speech levelRaise it if music wobbles in gaps; lower it if music never dips
RatioModerate, around 3:1 to 4:1Lower it if the dip sounds abrupt; raise it if the voice is still masked
AttackFast, under 10 msSlow it slightly if the dip clicks; speed it up if first words are buried
ReleaseMedium, 200 to 500 msLengthen it if music pumps between words; shorten it if music stays quiet too long
Reduction6 to 10 dB for podcastsLess for a quiet bed, more for a loud or dense track

Frequently asked questions

What is audio ducking in simple terms?

Audio ducking is a technique that automatically turns down one sound, usually music, while another sound, usually a voice, is playing. When the voice stops, the first sound returns to its normal level. It keeps speech clear without manual volume changes.

Is audio ducking the same as sidechain compression?

Not exactly. Sidechain compression is the most common method used to create ducking, but ducking is the result — one sound making way for another. You can also achieve ducking with volume automation or a dedicated ducking plugin, and sidechain compression is used for other effects too.

Do I need a DAW to use audio ducking?

You need audio software that supports sidechaining, ducking, or volume automation. Most DAWs support it, and some video editors and podcast tools include automatic ducking as a built-in feature. Check whether your software offers a sidechain input on its compressor.

How much should music dip under a voice?

For podcasts, a dip of around 6 to 10 dB usually works well. For video narration over a quiet bed, 3 to 6 dB may be enough. The right amount depends on how loud and dense the music is, so start conservatively and increase only until the voice is clearly intelligible.

Can audio ducking be used in live sound?

Yes. Live mixing consoles and broadcast processors often include ducking for announcements and commentary. A venue’s background music can dip automatically when a microphone is used, and a commentator’s voice can stay forward over crowd noise.

Does ducking work for music mixing too?

Yes, but more subtly. In a music mix, ducking can lower a pad, a reverb tail, or a competing instrument slightly when a vocal enters. The dip is usually smaller than in podcasting, because the goal is a balanced blend rather than a clear hierarchy between voice and music.

Conclusion

Audio ducking is a simple idea with a large effect: when someone speaks, the music steps aside; when they stop, it returns. Built on sidechain compression, it keeps voices intelligible in podcasts, videos, broadcasts and mixes without constant manual volume adjustments. The setup is straightforward — separate tracks, a compressor on the music, the voice as the sidechain source — and the settings that matter are threshold, ratio, attack and release. Start with gentle values, listen over a full section rather than a few seconds, and check the result on the device your audience actually uses. Done well, ducking is invisible; the listener simply hears every word and enjoys the music in the gaps.

Similar Posts

Leave a Reply

Your email address will not be published. Required fields are marked *