Find every silent gap in a recording and cut it out automatically. Set how quiet counts as silence and how long a pause has to be before it goes.
Drop your video here
MP4, MOV, WebM or MKV — up to 100MB
Free, no account needed, no watermark. Free tier: 1 file per day · max 10 minutes.
1 file per day · max 10 minutes
Example Output
The same 30-second recording after silent gaps above -30dB were cut: 30.04s became 21.27s, with 8.77 seconds of dead air removed — 29% shorter, and not one word lost.
How It Works
A podcast, webinar, tutorial, course lesson — anything with more dead air than you want.
Two numbers: how quiet counts as silence, and how long a quiet stretch has to last before it is cut.
The same recording with the gaps removed — usually noticeably shorter, and much faster to watch.
It scans the audio track for stretches below a volume threshold, and where one lasts longer than a minimum duration, it cuts that span out of both the audio and the picture. The result is the same content with the pauses squeezed out.
Two settings control everything. The threshold decides what counts as silence — measured in decibels below peak, because real recordings are never truly silent, they have room tone, air conditioning and breath. The minimum duration stops it butchering natural speech: without it, every gap between two words qualifies and the output sounds like a machine gun.
Unedited long-form recordings carry a surprising amount of dead air — thinking pauses, sips of water, waiting for a guest to unmute, restarts after a stumble. Cutting it is the single highest-yield edit on a raw recording, and it is entirely mechanical, which is why it should be automated.
Why Use This Tool
Every silent span in the file is found in one pass — no scrubbing the timeline looking for gaps.
Set the volume floor and the minimum gap length, because a quiet studio and a noisy room need different numbers.
An hour of raw recording routinely loses several minutes of dead air without a single word being cut.
Your original file is untouched. You get a new, tightened export.
Tuned for podcasts, webinars and tutorials rather than clips that are already tight.
Nothing is added to the picture.
The audio is analysed for spans quieter than a threshold you set. Any span longer than your minimum duration is cut from both audio and video, and the remaining pieces are joined.
Not with sane settings. The minimum-duration setting exists precisely to protect natural speech rhythm — set it too low and normal gaps between words get cut, which is what makes the audio sound chopped.
Around -30dB works for most recordings. A noisy room needs a higher (less strict) threshold so room tone is not mistaken for speech; a treated studio can go lower.
On unedited conversational recordings, commonly 5-15% of the total length. Tutorials and solo recordings with a lot of thinking pauses can be considerably more.
No. Filler words are speech, not silence, so they pass the volume test. This tool removes gaps, not words.
One file per day, up to ten minutes, on the free tier.
A tightened 55-minute podcast is still a 55-minute podcast. The question short-form asks is which 40 seconds of it deserve to exist on their own. ShortsHero reads the transcript, scores every moment, and cuts the ones that stand up alone.
Find the clips worth postingFree to start · No credit card required
Keep Going
Trim video online free. Set in and out points on a timeline and export just the section you want — no watermark, no sign-up, no software to install.
Upload a podcast episode and get a vertical, captioned, speaker-tracked clip back automatically. Built for turning long conversations into TikToks, Reels and Shorts.
Convert video to text free. Upload an MP4 and get an accurate, timestamped transcript with speaker labels where detectable. Download as TXT or SRT.
Enter your recording length and format to see how many short clips it realistically contains, how many editing hours that costs by hand, and the cost per clip.