Three underrated features in the podcast editors you already use
Bottom line: the overlooked podcast editing feature we would switch on first is Riverside asking which speaker to remove when voices overlap, because it solves the hardest cut in any interview. Close behind are Descript's per-word filler options and "Avoid harsh cuts", and Audacity 4's delete that closes the gap on every track.
"Underrated" is a judgement, so each entry says what the feature does, quoted from the vendor's own pages, and why we think people miss it. We make a competing editor, and none of the three entries is ours.
The list at a glance
| Feature | Where it lives | Why it is worth finding |
|---|---|---|
| Overlap-aware word removal | Riverside, all plans | Cut one speaker without taking the other |
| Per-word filler options, "Avoid harsh cuts" | Descript, English transcripts | Choose how each filler goes, skip the ones that would clip |
| Delete and close gap (all tracks) | Audacity 4, free | Multitrack cuts that stay in sync |
1. Riverside's overlap-aware word removal
Riverside's text-based editing is listed on "All plans": click a word or drag across several, press Delete, and the removed text stays "shown in gray with a strikethrough". The detail people miss is what happens when two people talk at once. When speakers overlap, Riverside asks which speaker to remove, rather than taking both. You can also mute words instead of cutting them.
Why it is underrated: you only meet it when two voices collide, which is exactly where cutting by transcript is hardest in an interview. Riverside's pages lead with recording, with tracks "recorded locally on each participant's computer or device"; this is the editing side, and it is easy to miss. More in our Riverside review.
2. Descript's per-word filler options and "Avoid harsh cuts"
Descript underlines common filler words in light blue, and its Remove filler words panel lists every instance with a timestamp and an audio preview. It is tempting to press the button and move on. The options are worth a look. Each filler can be:
- Deleted.
- Deleted and replaced with a gap, which keeps a pause where the filler was.
- Ignored, which "strikes through the filler word in the script but removes the audio".
- Removed from the transcript only, leaving the audio untouched.
Then there is the "Avoid harsh cuts" setting, which skips fillers that cannot be removed cleanly.
Why it is underrated: removing every filler can leave speech sounding clipped. Replacing some with a gap keeps the speaker's rhythm, and "Avoid harsh cuts" stops the tool cutting where it would damage the words around it. Two limits from Descript's own article: "This feature currently detects filler words in English transcripts only", and it uses AI credits (up to 30 per file each time). See our Descript review.
3. Audacity 4's "Delete and close gap (all tracks)"
Audacity 4 replaced Audacity 3's Sync-lock with ripple editing, documented in its transition guide as "Delete and close gap (all tracks)". You can also select several clips with Shift+Click.
Why it is underrated: if you edit a two-person recording on separate tracks, cutting a pause from one track and not the other is how the conversation drifts out of sync. Closing the gap on every track at once is the multitrack habit Audacity now builds in, and it is free, with "unlimited undo throughout the editing session". Our Audacity review covers the rest.
Why trust this list
Every claim above was read on the vendor's own pages on 17 October 2026: Riverside's help centre articles on removing words and on local recording; Descript's help centre article on removing filler words; and Audacity's 4 transition guide and Cut feature page.
We make Paraphrase, a competing editor, and none of the three entries is ours. Paraphrase leads on never generating speech and on showing every AI cut struck through for you to restore; it has none of the features above, since it has no multitrack, no speaker separation and no filler-word list. We have no customer reviews to cite.
What to do with this
Pick the tool you already pay for and try its entry on your next episode. There is a wider roundup of text-based editors if you are still choosing.
If you have tried all of these and cutting words is still the slow part, that is the one job Paraphrase does: you edit your own recording by its transcript, the AI cleanup proposes cuts you can see and undo, and nothing is ever generated. Make a free account on paraphrase.bot, no card needed, and try it on 15 minutes of your own audio.