How to Compress Audio in Audacity (File Size and Dynamics)

Audacity is the most-installed free audio editor on the planet, and it can "compress audio" two completely different ways. People searching for "how to compress audio in Audacity" usually want one of these but get lost trying to find it because Audacity does not disambiguate the terms. This guide covers both, in order, with the exact menu paths.

If you want smaller file size (the file is too big to email or upload), skip to the export section. If you want even loudness (vocals jumping around, podcast yo-yoing between whisper and shout), the Effect → Compressor section is what you want. The two have nothing in common except a shared word. For the full conceptual breakdown see audio compression explained.

Method 1: Compress File Size by Exporting at a Lower Bitrate

This is what most people actually want. Audacity itself stores audio as uncompressed 32-bit float PCM — internally everything is huge. The file size only goes down when you export to a compressed format like MP3, OGG, or Opus.

Step-by-step

1. Open your project in Audacity (or import the source file: File → Import → Audio). 2. Make any edits you need. Trim silence, normalize, fade — whatever. Editing happens before export. 3. Open the export dialog: File → Export Audio (Audacity 3.4+). On older versions: File → Export → Export as MP3 (or WAV, OGG, Opus). 4. Pick the destination folder and filename. 5. Choose Format: MP3 Files for the broadest compatibility. 6. Pick a Bit Rate Mode: Variable for best size-to-quality, Constant for predictable file size, Average for a compromise. Variable is the right default. 7. Pick a Quality preset (the dropdown labels these by VBR quality level — see the table below). 8. Choose Channel Mode: Stereo for music, Joint Stereo for tighter compression on most music, Mono for single-mic voice (halves the file size). 9. Click Export. 10. Fill in the Edit Metadata Tags dialog (or skip — click OK). 11. Audacity encodes and writes the file. A 5-minute song at LAME V2 (preset Standard) lands around 6–7 MB.

Audacity LAME quality preset reference

The dropdown values map to LAME presets. Use this table to pick the right one.

| Audacity label | LAME preset | Approx VBR average | Best for | |---|---|---|---| | Insane | --preset insane | 320 kbps CBR | Archival from a master, audiophile listening | | Extreme | -V 0 | ~245 kbps | Music for critical listening | | Standard | -V 2 | ~190 kbps | Default for most music — transparent on consumer gear | | Medium | -V 4 | ~165 kbps | Casual music listening, podcasts with music | | (custom) | -b 96 | 96 kbps | Voice podcasts, audiobooks (mono further halves it) |

For the deeper trade-off between VBR and CBR see VBR vs CBR MP3 and the bitrate ranges by use case in audio bitrate explained.

File-size math for typical inputs

A 10-minute stereo recording, sample rate 44.1 kHz:

• WAV (uncompressed): ~106 MB
• MP3 320 kbps CBR: ~24 MB
• MP3 192 kbps VBR (Standard): ~14 MB
• MP3 96 kbps mono: ~3.5 MB
• Opus 96 kbps: ~7 MB at noticeably better quality than MP3 96 kbps

Picking the right bitrate is the lever — encoder choice (MP3 vs Opus vs AAC) is secondary unless you are pushing below 96 kbps, where Opus pulls clearly ahead.

One catch: the LAME library

Audacity 3.4 and later bundle LAME automatically. Older Audacity versions (2.x) require downloading and pointing to lame_enc.dll (Windows) or libmp3lame.dylib (macOS) the first time you export MP3. If your version asks where LAME is, install the latest Audacity instead — saves the hassle.

Method 2: Dynamic-Range Compression with Effect → Compressor

This is the other compression — the one that makes a podcast voice sound consistent or evens out a vocal track. It does not change file size; it changes how the audio sounds.

Step-by-step

1. Open your project. Select the audio you want to compress (Ctrl/Cmd-A for the whole track). 2. Open the compressor: Effect → Volume and Compression → Compressor (Audacity 3.4+) or Effect → Compressor (older versions). 3. Set the parameters. For a starting point on spoken voice: - Threshold: -20 dB - Noise Floor: -40 dB - Ratio: 4:1 - Attack Time: 0.20 seconds (200 ms) - Release Time: 1.0 seconds - Make-up gain: checked (Audacity will auto-apply) 4. Click Preview to listen. Adjust threshold downward (more negative) to compress more, upward (less negative) to compress less. Adjust ratio higher for harder compression. 5. Click OK to apply. The waveform visibly flattens — peaks shrink, quieter parts come up. 6. After compression, run Effect → Volume and Compression → Loudness Normalization to set the final integrated loudness. -16 LUFS is a common podcast target (Apple Podcasts), -14 LUFS for music (Spotify).

Why fast attack is sometimes wrong

Audacity's compressor defaults to a ~200 ms attack, which is slow. For voice, that is fine — it preserves the natural feel of speech. For a snare drum or a pluck transient, that lets the peak through unchecked, which is sometimes what you want and sometimes not. Faster attack (~10 ms) clamps transients harder; slower attack (~50–100 ms) leaves them more natural. Adjust by ear.

Common Audacity compressor pitfalls

• Too much compression: ratio above 6:1 with a deep threshold makes voice sound squashed and unnatural. Back off.
• Pumping: short release (under 100 ms) causes audible level "breathing" between syllables. Lengthen the release.
• Noise floor floor too high: if you set noise floor above the actual recording floor, room noise gets pumped up between phrases. Set it 6 dB above your measured noise level.
• Forgetting normalization after: compression brings the average level up but the peak level down. You usually want a final pass of Loudness Normalization to hit your target LUFS.

When the Browser Is Faster Than Audacity

For pure file-size compression on a single file, Audacity is overkill:

• Install Audacity: ~150 MB download, several minutes to install.
• Open Audacity, import file: 30+ seconds for a long recording.
• Configure export, click through dialogs: 30 seconds.
• Encode and save: variable.

For a single MP3 you want smaller, /compress-mp3 takes around 10 seconds end-to-end with no install. The same is true for /compress-wav, /compress-m4a, /compress-ogg, and /compress-flac, or the format-agnostic /audio-compressor. Files are processed in your browser via WebAssembly — they never upload.

Audacity wins when:

• You also want to edit the audio (trim, fade, noise reduction) before compressing.
• You need dynamic-range compression (the other meaning), not file-size compression.
• You are batching tens of files with the same processing chain (use Audacity Macros).

The browser wins when you just need a smaller file in 10 seconds on whatever device you have.

Batch-Compressing in Audacity

Audacity supports batch operations through Tools → Macros. The setup:

1. Tools → Macros. 2. Click New to create a macro. Name it "Compress to MP3 V2." 3. Click Insert to add steps. For pure export: add ExportMP3 and configure the bitrate. 4. Save the macro. 5. Tools → Apply Macro → Files: pick the folder of input files. Audacity processes each one in turn.

This works for tens of files but slows down badly past ~100. For hundreds of files, FFmpeg in a shell loop is dramatically faster — see FFmpeg compress audio for the one-liners.

Quick reference: which method for which problem

| Problem | Method | Menu path | |---|---|---| | MP3 is too big to email | Method 1 (lower bitrate export) | File → Export Audio | | Podcast voice is jumpy | Method 2 (dynamic compressor) | Effect → Volume and Compression → Compressor | | Music master needs to sit at -14 LUFS | Loudness Normalization | Effect → Volume and Compression → Loudness Normalization | | Convert WAV to MP3 with one click | Method 1 (Export Audio) | File → Export Audio → MP3 | | Same processing on 50 files | Macros | Tools → Macros |

Verdict

For the most common search intent — "I have a recording and the file is too big" — Audacity's Method 1 works but is slow if you do not already have it installed and open. The browser tools at /audio-compressor are faster for single files. For dynamic-range compression on a podcast voice, Audacity's Effect → Compressor is genuinely the right tool and the price (free) is unbeatable.

For a fuller picture of what the word "compression" means in audio, see audio compression explained. For a comparison of browser vs desktop converters, see best free audio converter.