How to remove silence in Audacity with Truncate Silence

What Audacity's Truncate Silence effect does

Audacity is a free, open-source audio editor available for Mac, Windows and Linux. Its Truncate Silence effect scans the selected audio, finds every stretch that stays below a volume threshold for long enough to count as a pause, and shortens those gaps automatically. It is the fastest way to tighten dead air in a raw voice recording without dragging the playhead across the whole timeline by hand.

The effect works purely on amplitude. It does not understand words, so it cannot tell a deliberate dramatic pause from a moment where you were simply reaching for the mouse. That is why the two settings you feed it, the threshold and the minimum duration, decide almost everything about the result. Get them right and a rambling take becomes crisp; set them too aggressively and the audio starts to sound clipped and breathless.

Truncate Silence is best suited to spoken-word audio such as podcast episodes, voiceovers and audiobook chapters. It is not a mixing tool and it will not remove filler words like um or uh, because those are sounds, not silence.

How to remove silence in Audacity, step by step

Import your recording, then select the region you want to clean. Press Ctrl+A (Cmd+A on Mac) to select the entire track, or click and drag to limit the effect to one section. With a selection active, open the Effect menu, choose Special, and click Truncate Silence to bring up the dialog.

Set the Threshold first. This is the volume level below which audio counts as silence, adjustable roughly from -20 dB down to -80 dB. A value around -40 dB to -50 dB works for a clean voice recording; raise it toward -25 dB if quiet room tone is being left behind, or lower it if breaths are being swallowed. Then set the minimum Duration, the shortest pause the effect is allowed to touch. Around 0.3 to 0.5 seconds keeps natural sentence breaks while removing long dead air.

Finally choose an action and preview before you commit. Use the Preview button to sample a few seconds, adjust, and only click OK when the rhythm sounds right. Save an untouched copy of your project first, since Truncate Silence rewrites the waveform in place.

• Select audio (Cmd+A for the whole track)
• Effect > Special > Truncate Silence
• Set Threshold, then minimum Duration
• Pick Truncate or Compress
• Preview, then click OK

Truncate Detected Silence vs Compress Excess Silence

The dialog offers two ways to shorten each pause. Truncate Detected Silence, the default, cuts every qualifying gap down to a fixed target length that you type into the Truncate to box, in seconds. It is decisive and predictable: set it to 0.2 seconds and every long pause becomes a 0.2-second pause. This is ideal when you want a tight, high-energy edit and do not mind that the natural variation between pauses disappears.

Compress Excess Silence is gentler. Instead of a hard cap it shrinks each pause to a percentage of its original length, entered in the Compress to box. At 50 percent a two-second gap becomes one second and a shorter gap shrinks proportionally, so the recording keeps some of its original breathing rhythm while still losing the drag. Use Compress when a fully truncated edit sounds robotic or rushed.

A practical habit is to start with Compress at 30 to 50 percent for a natural feel, then switch to Truncate only if the pauses still feel too long. Whichever mode you pick, the Threshold and Duration settings above it stay in charge of what counts as silence in the first place.

The honest limits, and when to reach for a video tool

Truncate Silence is powerful but narrow. Its biggest limitation for creators is that Audacity is an audio editor: it edits the sound and nothing else. If your recording is a talking-head video, cutting the audio in Audacity does absolutely nothing to the picture, and re-importing the shortened audio leaves it drifting out of sync with the footage frame by frame. There is no way around this inside Audacity itself.

The settings are also fiddly. Because the effect is deterministic and blind to meaning, dialing in the right threshold for a noisy room can take several passes, and there is no preview of the whole finished result at once, only short samples. That trial-and-error is fine for a short voiceover and tedious for a long recording.

When you are editing video rather than pure audio, an automatic video silence remover is the better fit because it cuts audio and picture together and keeps them locked in sync. Cutting-Silence is a native Mac app that does exactly this: it detects silence on the audio track with an adjustable threshold, previews the entire edited video before you export, and runs 100% on your machine so nothing is uploaded. You get 5 full exports free, with analysis and preview always unlimited, so you can compare it against your Audacity workflow before deciding.