Recording Analyzer: a free, open-source audio metrics tool for Linux

[mcochris]

Recording Analyzer: a free, open-source audio metrics tool for Linux

Ever wanted to quickly check the dynamic range, noise floor, or EBU R128 loudness of your music files without loading files into specialized software? I've built a free, open-source Linux command-line tool - also WSL-compatible - that analyzes audio files for the metrics that matter most to objective listeners:

• Peak level (dBFS)
• Noise floor (dBFS)
• Crest factor (dB)
• Stereo correlation (phase)
• Loudness (EBU R128: LUFS, LRA, true peak)

Sample output:

$ reca.sh "Toccata and Fugue in D Minor.flac"

Audio Analysis: "Toccata and Fugue in D Minor.flac"
===================================================

Left Channel:
  Peak Level:     -1.94 dBFS
  Noise Floor:    -inf dBFS
  Crest Factor:   8.91 dB

Right Channel:
  Peak Level:     -2.21 dBFS
  Noise Floor:    -inf dBFS
  Crest Factor:   8.06 dB

Stereo Correlation:
  Average Phase:  0.15

Loudness (EBU R128):
  Integrated Loudness:  -16.20 LUFS
  True Peak:            -1.94 dBTP
  Loudness Range:       18.60 LU

Other features:

• Analyze single files or entire directories and subdirectories
• Display file metadata: genre, artist, album, track, duration, year, sample rate, average bit rate, bits per sample
• Upload your JSON output to recording-analyzer.mcochris.com for an interactive sortable table, spreadsheet download, and custom playlist builder — filter by crest factor, loudness, or any other metric
• Access music files on network-attached storage devices
• Includes a verification script that cross-validates results against SoX and other tools
• Your data stays yours: no ads, no personal data collected, no tracking cookies
• Open source, permissive MIT license

File formats supported: aac, ac3, aif, aiff, amr, caf, flac, m4a, mp3, ogg, opus, pcm, wav, wma

Windows users: requires WSL, available free from the Microsoft Store.

GitHub: program info and installation

If you find it useful or run into any issues, I'd love to hear from you here or via a GitHub issue. I'm particularly curious whether the metrics match what you see in your own tools.