How to use Audio Metadata Viewer
- 1Drop in an audio file
- 2Read the stored tags — title, artist, album, BPM, comments
- 3See the embedded cover art if there is one
- 4Check the measured loudness figures below the tags
- 5Nothing is changed; the file is only read
What Audio Metadata Viewer does
An audio file carries two very different kinds of information. Tags are text written into the file by whoever created or edited it — title, artist, album, comments — and they can say anything, because nothing verifies them against the audio.
The second kind cannot be written at all: it has to be measured by reading every sample. Peak level, average level, how many samples were clipped, whether the waveform sits off-centre. Those figures describe what the audio actually is rather than what someone typed about it, and they are what tell you whether a recording is usable.
When to use it
- Diagnosing a recording that sounds wrong
- A clipped sample count above zero explains distortion immediately, and does so before you spend time trying to fix it with volume adjustments that cannot help.
- Deciding how much you can raise the level
- The peak figure tells you exactly how much headroom is left. A file peaking at −12 dB can safely come up 11 dB; one at −0.5 dB cannot come up at all.
- Checking a download before using it
- Tags reveal where a file came from and what wrote it, which is useful when a track has been through several hands.
- Understanding why a track sounds flat
- A dynamic range of only a few decibels means the recording has been heavily compressed — loud throughout, with no light and shade.
Audio Metadata Viewer settings explained
- Peak level
- The loudest single sample, in decibels below maximum. This is the number that decides how much gain a file can take before it clips.
- Average level (RMS)
- The overall energy of the recording rather than its loudest instant. Two files with identical peaks can differ enormously here, and RMS is much closer to how loud something actually sounds.
- Dynamic range
- Peak minus average. A large figure means quiet passages and loud ones; a small figure means the recording is loud throughout. Heavily mastered modern music often sits below 8 dB.
- Clipped samples
- How many samples sit at the digital maximum, and what share of the total that is. Any figure above zero means distortion is already baked into the file and cannot be removed.
- DC offset
- Whether the waveform is centred on zero. An offset wastes headroom and can produce a click at the start and end. It usually indicates a fault in the recording hardware.
- Leading silence
- Reported when there is more than a twentieth of a second of quiet at the start — the thing you would trim first, and easy to miss by eye on a long waveform.
Tags compared with measurements
| Tags | Loudness measurements | |
|---|---|---|
| Source | Written into the file | Calculated from every sample |
| Can be wrong | Yes — it is just text | No — it is what the audio is |
| Survives conversion | Often lost | Recalculated each time |
| Tells you | What it claims to be | Whether it is usable |
Limits and what to expect
- This reads and measures only — it does not edit tags or alter the audio.
- Tags absent from the file cannot be inferred; a missing artist stays missing.
- Measurements cover the whole file, so a single loud moment shapes the peak and dynamic range figures.
- Loudness is reported as peak and RMS rather than LUFS, which is the broadcast standard.
- 150 MB per file; every sample has to be read, so long recordings take a moment.
Troubleshooting
- All the tag fields are empty
- The file carries no metadata, which is normal for recordings from a phone, a dictaphone or an audio editor's export. Tags are added by music software and distributors, not by recording devices.
- Clipped samples shows a number above zero
- The recording was made or processed too loud and the peaks were flattened. That distortion is permanent — lowering the volume now makes it quieter but does not restore the waveform. The recording has to be redone to fix it properly.
- The dynamic range is very small
- The audio has been heavily compressed, deliberately in most commercial music. It is loud throughout with little variation. Nothing is wrong with the file; it is a production choice you cannot undo.
- There is a DC offset
- The waveform is not centred on zero, usually a fault in the recording hardware. It wastes headroom and can cause clicks. Most audio editors have a DC offset removal function; this tool reports it but does not correct it.
Your files and your privacy
The file is decoded and analysed inside your browser, and nothing is uploaded. That is the point for a tool of this kind: something that tells you what a file discloses should not itself be the thing that discloses it. Your file is not modified — this reads only.
What to use next
FAQ
What is the difference between peak and RMS?
Peak is the single loudest sample; RMS is the average energy across the recording. Peak decides how much you can raise the level before clipping, RMS is much closer to how loud something actually sounds. Two files with the same peak can differ enormously in perceived loudness.
What does a clipped sample count mean?
It counts samples sitting at the digital maximum, where the waveform has been flattened rather than reproduced. Any number above zero means distortion is already in the file permanently. Reducing the volume afterwards makes it quieter but cannot restore what was flattened.
What is a good dynamic range?
It depends entirely on the material. Classical recordings often exceed 20 dB; heavily mastered pop frequently sits below 8. A low figure is not a fault, it is a production decision — but it does mean the recording is loud throughout with little variation.
Why are my tags empty?
Because nothing wrote any. Phones, dictaphones and audio editors generally export without metadata; tags are added by music software, distributors and tagging tools. An empty tag block is normal for a recording you made yourself.
Can I edit the tags here?
No, this reads only and never modifies your file. Writing tags is a different operation — the MP3 converter can attach title, artist and album when it exports.
Is my audio uploaded?
No. Decoding and analysis both happen inside your browser on your own device, with no request to any server.