- Home
- Skills
- Workflow Automation
- Dead Air Cutter — Local
Works with the AI Agents you already use
Dead Air Cutter — Local
Remove silences and dead air from video and audio files locally using ffmpeg and Node.js.
$5
Dead Air Cutter — Local
Example session with this skill installed
Cut the pauses out of my voiceover "vo.wav" — normal style.
- Read your context and instructions
- Compiled the dead air cutter
Done — vo_cut.m4a is next to the original, which is untouched. Nothing left your machine: ffmpeg did all the listening and cutting locally.
What I did
- Previewed first with
--dry. It's free and instant, and writes nothing. - Cut with the
normalstyle, which keeps 0.15 s of air around every phrase.
normal (noise -35 dB, pauses ≥ 0.5 s): 11 pause(s), 37.7 s → 30.2 s (−20 %)
vo_cut.m4a
Result
| Source | vo.wav (spoken voiceover, 37.7 s) |
| Output | vo_cut.m4a (audio in, audio out) |
| Pauses cut | 11 |
| Length | 37.7 s → 30.2 s (−20 %) |
| Network calls | none |
| Cost | $0 per file |
Other styles on the same file
All three come from the same instant preview:
| Style | Air kept around speech | Length |
|---|---|---|
| gentle | 0.3 s | 33.3 s (−12 %) |
| normal | 0.15 s | 30.2 s (−20 %) |
| jump | 0 s | 27.0 s (−28 %) |
Next steps
- Want it tighter? Say "jump" and I'll re-cut. It overwrites
vo_cut.m4a. - In a noisy room the pauses may not count as silence. If too few get cut, I'll retry with
--noise -30.
Connects securely to your tools. The creator never sees your data.
About this skill
Manually editing out pauses in vlogs, podcasts, or lectures is a tedious, repetitive task. This skill automates silence removal locally using ffmpeg, tightening your recordings without requiring cloud APIs, accounts, or data uploads.
What it does
- Detects silence by analyzing audio loudness via ffmpeg silencedetect.
- Removes dead air from video or audio files while preserving speech.
- Supports multiple styles ranging from natural interview pacing to aggressive YouTube jump cuts.
- Maintains quality by re-encoding video to H.264 CRF 18 with AAC audio.
- Previews cuts with a dry-run mode that calculates time savings without writing files.
How it works
- Selection: You provide the path to a video or audio file and choose a cutting style (gentle, normal, or jump).
- Preview: The agent runs a dry run to report how many pauses were found and the projected final duration.
- Execution: The agent executes the ffmpeg filter chain to generate a new file with the
_cutsuffix. - Validation: The agent reports the final file path and the total time saved.
Frameworks & tools
This tool requires Node.js 20+ and ffmpeg (with ffprobe) installed on the local system PATH. It utilizes a local JavaScript script to interface with the ffmpeg CLI.
Why this beats prompting it yourself
Writing complex ffmpeg filter chains for silence detection is error-prone and requires precise threshold tuning. This skill handles the math for different styles and manages temporary filter files, ensuring you get clean cuts without clipping word endings.
Use cases
- Vlog editing to create snappy jump cuts for YouTube content.
- Podcast tightening by removing long gaps in raw interview recordings.
- Lecture cleanup to eliminate dead air during whiteboard transitions.
- Voiceover polish by stripping out breaths and pauses between takes.
Known limitations
Only processes the first audio track. Background music may prevent silence detection. Does not generate subtitles or carry over existing subtitle tracks. Re-encoding large files takes time.
Requires ffmpeg/ffprobe on PATH. Only processes the first audio track. Background music interferes with silence detection. Overwrites existing _cut files.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 27 days ago
Requires