Audacity: what it is and how to start

A free audio recorder and editor that captures from a microphone, cuts and cleans recordings, and exports to every common format.

Who it is for

Podcasters, voiceover recorders, and anyone who needs to clean up spoken audio. Less suited to music production, which needs multi-track recording and virtual instruments that Audacity is not built around.

What it costs

Completely free and open source under the GNU General Public License, with no paid tier, no watermark and no limits. Desktop only — there is no mobile version.

Quick facts

Category: Music Production & Audio — Audio Recording & Editing
Pricing model: free
Runs on: Windows 10 and 11, macOS 12 and 13, Linux
Difficulty for a beginner: beginner
Official site: https://www.audacityteam.org

How to get started with Audacity

Fix your room and microphone distance first: Nothing in the software fixes echo or a distant microphone. Record somewhere with soft furnishings and sit roughly a hand span away. This matters more than every setting in Audacity combined.
Select the right input device before recording: Audacity does not always pick your USB microphone automatically and will happily record from the laptop's built-in one. Check the device dropdown, then record ten seconds and listen back before committing to a full take.
Set levels so peaks stay below maximum: Watch the meter while speaking normally and aim for peaks comfortably short of the top. Anything that hits the ceiling is clipped and permanently distorted — that damage cannot be repaired afterwards.
Record a few seconds of silence: Before you start talking, capture the empty room. Noise reduction needs a sample of that background to remove it, and without one the tool cannot work properly.
Apply noise reduction using that sample: Select the silence, open Effect then Noise Reduction, and click Get Noise Profile. Then select everything and apply it. Use a light setting: aggressive noise reduction makes voices sound underwater, which is worse than the hiss.
Cut the mistakes and tighten the pauses: Select and delete. Most of editing spoken audio is removing false starts, filler words and long gaps. Leave some natural breathing, since removing every pause sounds unsettling.
Even out the volume, then export: Use Effect then Normalize or Compressor so the level stays consistent rather than loud. Export as WAV for a master you keep, and MP3 for publishing. Always export MP3 from the project rather than from another MP3.

Should you use this instead of something else?

Descript and similar tools edit audio by editing a transcript, which is dramatically faster for long interviews and costs money. GarageBand is free on Apple hardware and better for music, since it is built for multi-track work and instruments. Audacity wins on being free, open source, cross-platform and permanent — there is no account, no subscription and nothing that can be taken away.

Common beginner mistakes

Recording from the laptop's built-in microphone because the input device was never changed.
Recording too quietly to avoid clipping, then amplifying afterwards, which raises the hiss along with the voice.
Applying heavy noise reduction, producing a hollow underwater voice that is more distracting than the original background noise.
Editing out every pause, leaving delivery that is breathless and hard to follow.
Exporting MP3 from a previously exported MP3, stacking compression damage each time.
Not saving the project file, so a single later correction means redoing the whole edit.

Last verified: 2026-08-07. Source: https://www.audacityteam.org/about/

Read the full Music Production & Audio guide for how this fits into the wider process.