Podcast Trimmer

It is built for one job: taking a finished recording and getting it ready to publish. That means cutting the seconds before you started talking and after you stopped, pulling out the clips you want, and landing the whole thing at the loudness a podcast platform expects — without the file ever leaving your device.

• A trim that changes nothing it does not have to. Cutting an MP3 and saving it as an MP3 copies the encoded audio across untouched. No decode, no re-encode, no generation loss — and it takes milliseconds regardless of how long the episode is.
• Dead air at both ends, found for you. The tool listens for where the talking actually starts and stops and moves the handles there. You can see exactly how much it wants to cut before anything happens.
• Publishing loudness, exactly. Pick -16 LUFS and the file lands at -16 LUFS. The true-peak ceiling is held with a limiter, and if peaky audio forces the level lower, the tool tells you the number it actually reached instead of the one you asked for.
• Several clips from one episode, in one pass. Mark the moments you want as separate clips and export them together as a ZIP — a show-notes chapter set, or three pull-quotes for social, without loading the episode again.
• Six output formats. MP3, M4A, AAC, OGG, FLAC and WAV. Lossless formats carry the source's own sample rate, channel count and bit depth rather than quietly flattening them.

Audio in an MP3 or an M4A is already compressed. A tool that opens it, cuts it and saves it again has to decode the audio back to raw samples and then compress it a second time — and every one of those round trips throws away a little more. Do it three or four times across an editing session and it is audible.

Cutting does not actually require any of that. The compressed audio is stored in small frames, and a cut that lands on a frame boundary is a matter of copying the frames you want into a new file. That is what this tool does whenever it can: the audio you download is byte-for-byte the audio that was in your episode. Measured on a ten-minute episode, copying four minutes out of it takes 28 milliseconds against three and a half seconds to re-encode the same material.

When it happens

• The output format is the same as the input format.
• No fade is set.
• No loudness target is set.
• The sample rate, channel count and bit depth are left as the source's.
• Dead-air trimming is welcome — it only moves where the cut lands, so it does not change a single sample.

The one trade-off

A copied cut lands on the nearest frame boundary rather than on the exact millisecond you asked for — for MP3 that is about 26 milliseconds. If you need a cut placed to the sample, choose WAV or FLAC and the tool will re-encode losslessly instead. The panel beside the controls always says which one you are getting, and why.

Almost every recording starts with the gap between pressing record and starting to speak, and ends with the gap between finishing and reaching the button. They are the two easiest seconds to remove and the two most often left in.

Turn on dead-air trimming and the tool scans the episode for silence, finds the first stretch at the very beginning and the last stretch at the very end, and moves the trim handles to where the audio really starts and stops. It shows you how much it is proposing to cut before you commit, and you can drag the handles afterwards if it was too eager.

If it cuts too much or too little

• Raise the threshold if a noisy room is being counted as speech, so more of the start gets cut.
• Lower it if a quiet first word is being cut off.
• Raise the minimum gap length if a breath before the first word is being treated as the start of the episode.

What it deliberately will not do

It never touches the pauses inside the episode. Removing those changes the rhythm of a conversation and cuts breaths, and it needs controls per gap to be usable at all — which is a different tool. If that is what you are after, the silence remover does it properly.

Podcast apps normalise what you upload. If your episode is quieter than the platform's target it gets turned up, and anything you did to control the dynamics gets undone; if it is louder, it gets turned down, and the peaks you thought you had headroom for may not survive. Delivering at the target in the first place is how you keep control.

-16 LUFS with a true peak no higher than -1 dBTP is Apple's published recommendation for podcasts and the figure most of the industry has settled on. Choose it here and the exported file lands there: the tool measures the exact clip you are exporting, works out the difference, and applies it.

The targets

• -16 LUFS — Apple Podcasts, Spotify, and most podcast apps.
• -19 LUFS — the older stereo convention, still used by several networks.
• -23 LUFS — EBU R128, the target a broadcaster will ask for.
• Off — keep the original level, which is also what keeps a trim lossless.

When the ceiling wins

A true-peak ceiling is a hard limit and a loudness target is a target. On audio with very sharp peaks, holding the peak below -1 dBTP means the overall level cannot reach the target as well. When that happens the tool holds the ceiling, reports the loudness it actually achieved, and says why — rather than displaying the number you asked for over a file that does not match it.

A finished episode is usually the source for more than one thing: the episode itself, a couple of pull-quotes for social, maybe a chaptered set for the show notes. Marking those as separate clips and exporting them together means loading the episode once instead of four times, and every clip gets the same fades, the same format and the same loudness treatment.

Each clip becomes its own file, and the whole set downloads as a single ZIP with the filenames already numbered. Free exports one clip at a time; Pro pulls up to twenty out of a single episode in one pass.

• MP3 The safe default. Every podcast host and every app takes it, and a trim of an MP3 into an MP3 is the lossless copy path.
• M4A AAC in an MP4 container. Better quality per kilobyte than MP3 at the same bitrate, and the format Apple's own tools prefer.
• AAC The same audio as M4A in a raw stream rather than a container. Choose it only if something downstream specifically asks for it.
• OGG An open container, useful where MP3 licensing is a concern. Less widely supported by podcast apps.
• FLAC Lossless and roughly half the size of WAV. The right choice for an archive master or for handing audio to someone else to work on.
• WAV Uncompressed. Large, universally readable, and the format to pick when a cut has to land on an exact sample rather than a frame boundary.

Everything here runs on your own device. The file is never uploaded, never queued on a server and never held anywhere after you close the tab — which also means there is no upload wait, so a large episode starts being useful immediately instead of after a progress bar.

That matters more for podcast audio than for most things people trim. An unreleased episode, an interview under embargo, a recording with a guest's personal details in it — none of those should be sitting in someone else's queue, and here they are not.

Measured on a ten-minute episode: a same-format trim copies in 28 ms, a full re-encode takes 3.5 s, and a re-encode with a loudness target takes 6.3 s. Dead-air detection over the whole episode costs about 0.6 s.

• Remove silence throughout
• Audio cutter
• Audio normalizer
• Audio compressor
• Audio converter