sxsphinxstack

Skills / Record clean audio

Record clean audio

Get clean voice audio from the gear they already own. Room, mic distance, levels, and a cleanup pass in Audacity. Use when they say "my audio sounds bad", "how do I record my voice", or before any recording session.

Use this skill. Nothing to install.

Open in ChatGPT Open in Claude prefilled and ready; just hit send
for Gemini, Copilot, or Cursor: paste it, then say "use the record clean audio skill"

Don't have an agent?  ·  Raw file: skills/record-clean-audio.md

Watch the first two minutes

This is how a session goes, including the worry nobody says out loud. Click through it.

(pastes the skill) my recordings sound terrible and honestly i hate my voice
Two different problems, and only one is real. The harsh echoey sound is your room, and a blanket fixes it today. Hating your recorded voice is near universal: you normally hear yourself through your own skull, so the recording sounds like a stranger. Everyone else already hears the recording version.
so i don't need to buy a mic?
No purchase needed. Your phone's bottom mic at a hand-span, or wired earbuds clipped near your collar, gets a clean result. What do you actually need this audio for? We record the real thing today, so pick something you'll use.
a voiceover for a tutorial i'm making
Good. Room first: clap once, right now. If it rings, we move to the deadest spot you have: a closet full of clothes, a carpeted bedroom, even a parked car. Then fans off, notifications off on every device, and record ten seconds of "silence" so we can hear what's left.
ok the silence recording has a hum in it
That's the AC or the fridge; kill it for the take if you can, because whatever you hear now will be in every take. Anything that has to stay, Audacity's noise reduction removes with the room-tone clip you just made. It's free, and I'll walk you through the exact settings after we record.

Scripted example of a real session.

Don't I need to buy a microphone?

No, and this is the point of the skill. Most bad recordings are bad rooms: hard walls, running fans, a mic at laptop distance. Fix those three things, which cost nothing, and a phone's bottom mic at a hand-span produces audio most listeners will call professional. The skill runs the whole setup with you: the clap test to pick the room, a duvet over a chair if the room needs deadening, a real level check on your actual first lines, and a cleanup pass in Audacity, which is free on every platform. The session ends with a genuine take of something you need, exported, because a test recording you never use teaches you nothing.

What you end up with

A clean exported take plus the session sheet that made it, so you can repeat the setup alone next time.

Example
SESSION SHEET — TUTORIAL VOICEOVER
Gear: phone, bottom mic  ·  Room: bedroom closet, door closed
SETUP 1
Clap test: hallway rang, closet passed
Noise: AC off, phone on do-not-disturb, window shut
Distance: a hand-span (15–20 cm), talking past the mic, slightly off-axis
Levels: real first lines at performance volume, peaks at −12 to −6 dB
Room tone: 10 seconds of silence recorded in position
CLEANUP IN AUDACITY, IN ORDER 2
1. Noise Reduction from the room-tone profile, about 12 dB, gentle
2. Cut false starts, long gaps, mouth noises; natural pauses stay
3. Loudness Normalization to −16 LUFS
FILES 3
voiceover-take3.wav (for the video edit) · voiceover-take3.mp3 (192 kbps) · Audacity project kept
  1. The room did most of the work. Everything in the setup block is free, and it removes more noise than any plugin applied afterward.
  2. Order matters. Noise reduction needs the room tone first, and aggressive settings sound underwater, so the skill starts gentle and listens back with you.
  3. Two exports on purpose. WAV feeds further editing without quality loss; MP3 goes anywhere directly. The project file means the next fix takes minutes.

Questions people actually ask

Do I really not need any equipment?

Really. A phone or wired earbuds with an inline mic, plus a quiet soft room, produces a usable result. The skill is built around the gear you already own, and it will not suggest purchases.

Why do I hate the sound of my own voice?

You normally hear your voice partly through bone conduction in your own head, which adds depth a recording does not have. The recorded version is the one everyone else has always heard. It fades with exposure, and clean audio makes it fade faster.

What is room tone and why record it?

Ten seconds of "silence" recorded in position. Audacity's Noise Reduction learns the room's background hiss and hum from that clip, then subtracts it from the whole take. Without it, the cleanup pass has nothing to work from.

Can you fix a recording I already made?

Partly. If it contains a stretch of silence, noise reduction can pull out steady background hum. Echo from a hard room is baked in, though, and re-recording in a treated spot is usually faster than fighting it.

What settings do I export with?

WAV for anything you will edit further, MP3 at 192 kbps or higher for anything going straight out, and Loudness Normalization to about −16 LUFS first, a standard spoken-word level that sounds right on phones.

Where to go from here

This skill is the audio floor for everything else you record:

Podcast episode Screen-recording tutorial Video resume

Need something worth recording? These projects are built on a voice and a phone:

Grandparent interview Best-friend episode Street oral history
Curious? Read the full skill — the exact instructions your agent gets
---
name: record-clean-audio
category: media
description: Get clean voice audio from the gear they already own. Room, mic distance, levels, and a cleanup pass in Audacity. Use when they say "my audio sounds bad", "how do I record my voice", or before any recording session.
---

# record-clean-audio

Help someone record voice audio that sounds clean, using
what they already own — a phone, earbuds with a mic, or a laptop.
No purchase is required for a usable result. The session ends with a
real recording made, cleaned, and exported, so run a genuine take of
whatever they actually need the audio for.

## The room beats the mic

Most bad recordings are bad rooms, not bad mics. Before touching
settings:

- Pick the deadest room they have: closet full of clothes, carpeted
  bedroom, car parked with the engine off. Avoid kitchens, bathrooms,
  and any room that echoes when they clap once. Do the clap test
  together.
- Turn off noise sources: fans, AC, open windows, notifications on every
  device in the room. Record 10 seconds of "silence" and listen back
  loud — whatever you hear now will be in every take.
- Soft stuff absorbs: a duvet over a chair behind the laptop, facing
  a wardrobe, or recording under a blanket. Judge the setup by the test audio.

## Mic and levels

- Distance: a hand-span from mouth to mic (15–20 cm). Phone mic
  works fine at this range; the bottom mic of the phone is the good
  one. Wired earbuds with an inline mic are also solid — clip or
  hold the mic near the collar.
- Slightly off-axis: talk past the mic, not straight into it, to
  dodge plosives (the "p" pops). No pop filter needed at an angle.
- Set levels with a real test: have them read their actual first
  lines at performance volume. Peaks should sit around -12 to -6 dB
  in the meter — loud but never touching the top. On a phone, use a
  free recorder app that shows a meter (or record a test and check
  it in Audacity).
- Record a room-tone clip: 10 seconds of silence in position. The
  cleanup pass needs it.

## Cleanup in Audacity

Audacity is free on every platform. Import the recording, then in
this order:

1. Noise Reduction: select the room-tone section, Get Noise
   Profile, then apply to the whole track with gentle settings
   (around 12 dB reduction). Aggressive settings sound underwater —
   listen and back off if voices go metallic.
2. Cut the mistakes: false starts, long gaps, mouth noises. Keep
   natural pauses.
3. Normalize or Loudness Normalization to about -16 LUFS for spoken
   word — a standard level that sounds right on phones.
4. Export WAV for anything they will edit further, MP3 (192 kbps+)
   for anything going straight out.

## Done

- A real take of their actual content, recorded in the treated spot
- Before/after comparison played back — they hear what the room and
  the cleanup each did
- Exported file saved somewhere they can find it, plus the Audacity
  project if more editing is coming

This skill feeds podcast-episode, video-resume, and
screen-recording-tutorial — the audio bar is the same everywhere.