Pad the ends
Lead-in and tail-out add silence before and after the audio without touching a sample of it. The result is one file, in the format the source arrived in, with a longer duration and the same content.
Trailing silence is the one most often required. A voiceover that ends on the last consonant sounds cut off, and a hard stop leaves no room for a player’s own fade. One to two seconds is the usual ask, and it is the default here.
Leading silence is more often a timing fix: lining a stem up against picture, or holding a track back so an announcer can talk over the top.
Insert at a point
Scrub to a spot, stop, and set an insert length. The file is split there, the silence goes in, and the two halves are written either side of it. Useful for opening a hole for a sound effect or stretching a pause that was too short to work in the edit.
That seam is the one place the audio is touched. A 5 ms blend runs at each edge of the cut, because joining a waveform mid-cycle to a flat line steps the signal, and a step is an audible click. Five milliseconds is far too short to hear as a fade and long enough to remove the step.
Silence is not the absence of a track
The output is real audio: samples at zero, at the source’s own sample rate and channel count. Players show the longer duration, editors show a flat waveform, and a loudness meter reads it as part of the programme.
That last point matters for delivery. Integrated loudness to ITU-R BS.1770-4 gates out silence rather than averaging it in, so padding does not drag a measurement down. Check it with the loudness meter if a platform’s target is the reason for the padding.
Going the other way
To take gaps out rather than put them in, use the silence remover. To cut the file at its gaps into separate tracks, use the silence splitter.