How to Make a Clean Version of a Song (a.k.a. Clean Edit)

What "clean version" means

A clean version is a copy of a song with the explicit words removed — usually muted, bleeped, or pitched out. The melody, beat, runtime, and arrangement stay identical. Only the swears change.

"Clean version" and clean edit are interchangeable in practice; you'll see both terms used by labels, DJs, and streaming services.

Some people also use radio edit, which technically means clean lyrics plus a shorter runtime trimmed for broadcast (usually 3:00–3:30). If you only need a swear-free version of a song without changing its length, "clean version" or "clean edit" is the term you're looking for. We cover the radio-edit difference in a section below.

You'd want one for weddings, corporate events, gym and fitness classes, content creation, Twitch, TikTok and YouTube streams, school dances, family gatherings, kids' playlists, podcasts, public-event PA systems — anywhere explicit lyrics would be a problem.

Step 0: Check if an official clean version already exists

Before you spend time editing, search for the song on Spotify, Apple Music, or Tidal and look for a version tagged "Clean." Major-label releases — most pop singles, post-2015 hip-hop, and anything from a top-40 artist — almost always have an officially-mastered clean version uploaded alongside the explicit one.

How to find it:

• Spotify: Search the song name plus "clean" or scroll the artist's discography. The clean version appears as a separate track with a [Clean] tag in the title or as a separate album entry.
• Apple Music / iTunes: Same approach — the album page lists both versions side by side, and the explicit one carries the red [E] badge.
• Tidal / Amazon Music: Similar tagging; check the alternate-versions section of the song page.

An official clean version was mastered by the label's engineers, often with creative replacements (reverse, pitch-shift, alternate takes) instead of dead silence. It will always sound better than anything you produce yourself, so use it if it exists.

When the official clean version doesn't exist:

• Pre-2010 hip-hop and rap — labels often didn't bother
• Indie and underground artists — no label pressure to release one
• Artists who skip them on principle — some major artists have refused on certain releases as a creative statement
• Soundtrack-only or deluxe-edition tracks

If no clean version exists, keep reading.

Method 1: DIY in Audacity (free)

Audacity is free, runs on Mac, Windows, and Linux, and is the most common tool for one-off clean edits. Here's the workflow:

1. Install Audacity from audacityteam.org. Open your song file (File → Open).
2. Find the timestamps of every explicit word. Pull up the song's lyrics on Genius or Musixmatch in another tab. Play through the song and note the time of every word you need to mute — Audacity displays the playhead time in seconds along the top.
3. Select the explicit word. Click and drag on the waveform to highlight just that word. Zoom in with View → Zoom → Zoom In for accuracy — explicit words are often only 200–400 ms long, and selecting too much eats nearby vocals.
4. Apply silence with crossfades. Generate → Silence will replace the selection with dead air, but a hard mute creates an audible click at the boundary. Instead, apply Effect → Fade Out, then Effect → Fade In. This creates a smooth crossfade in and out of silence (roughly 30–50 ms on each side). Alternative: Effect → Amplify to -50 dB instead of full silence — sometimes preserves rhythm better for fast-rapped sections.
5. Watch out for the "f" trap. Fricative consonants (the "f" in "fuck," the "sh" in "shit") start ~50–100 ms before the visible vowel. If you only mute the loud vowel sound, the consonant slips through and the word is still audible. Always extend your selection 50 ms to the left of where you think the word starts.
6. Repeat for every explicit word. A typical hip-hop track has 10–30 explicit words. Budget 1–2 minutes per word once you're warmed up.
7. Export. File → Export → Export as MP3. Choose 320 kbps for best quality — the default 128 kbps degrades the music noticeably.

Realistic time estimate: 20–40 minutes per song for a clean result. Add another 20 minutes if it's your first time using Audacity.

Method 2: Logic, Pro Tools, or Audition

If you already work in a DAW, the workflow is identical in concept but faster in execution because of automation tracks:

1. Drop the song on an audio track.
2. Use a volume automation lane to drop the gain to silence over each explicit word, with 30–50 ms ramps on each side.
3. Logic users: enable track automation, then drag the volume node down to silent with curves on each side. Pro Tools: same idea via the "vol" automation lane. Audition: use the Mute effect with key-frame envelopes.
4. Bounce the track to MP3 or WAV when done (320 kbps MP3, or 16-bit / 44.1 kHz WAV for archival).

The advantage over Audacity is that automation lanes are non-destructive — you can tweak the cut points after listening to the full song without redoing work. Disadvantage: per-song time is similar (15–30 minutes), and you need a DAW license.

Method 3: Hire a freelancer

If you only need one or two songs cleaned and don't want to learn Audacity, post a job on Fiverr ("clean edit song") or Upwork. Typical pricing: $5–25 per song with 1–2 day turnaround.

Worth it when: it's a wedding playlist, a one-time event, or a particularly complex track with layered vocals, ad-libs, or background swears.

Not worth it when: you have more than ~5 songs to clean — the cost adds up fast and you still wait 1–2 days each.

Method 4: Use CENSORLY (automated, ~30 seconds)

CENSORLY does the same job as the Audacity workflow above, automatically:

1. Upload your song (drag and drop). MP3, WAV, FLAC, and M4A all work.
2. CENSORLY transcribes the vocals with word-level timestamps using AI speech recognition.
3. Every explicit word is automatically detected and muted with crossfades.
4. Review the auto-detection in a clickable transcript — toggle individual words back on or off, or drag custom mute regions on the waveform for anything the AI missed.
5. Download a 320 kbps MP3 with original metadata and album art preserved.

Time per song: ~30 seconds of compute, plus however long you spend reviewing — usually under a minute.

What CENSORLY doesn't do: it doesn't shorten songs, doesn't replace explicit words with creative substitutes (reverse, pitch-shift, sound effects), and doesn't generate alternate vocal takes. If you need any of those, you'll do them in a DAW. For straight muting of explicit words — the actual job 95% of the time — CENSORLY handles it end to end.

Pricing: First song is free. After that, plans start at $9/month for 10 songs, with one-time credit packs available if you don't want a subscription.

Try CENSORLY with your first song free →

What about "radio edits"?

A radio edit is technically two things stacked:

1. Clean lyrics — what we covered above.
2. Shorter runtime — usually trimmed to 3:00–3:30 to fit broadcast rotation slots, sometimes with intro tightening so vocals come in by 0:15.

If you searched for "how to make a radio edit" and only need the clean part, any of the methods above will get you there. Most artists do these as separate passes anyway — clean the lyrics first, then trim the structure in a DAW.

If you also need the length cut, that's structural editing — drop a verse, shorten an instrumental break, tighten a bridge — and it's a DAW job (Audacity, Logic, Pro Tools). Use CENSORLY for the clean step, then handle the trim separately.

Frequently asked questions