Audio Analysis & Metadata

What to look for when checking an audio file

Audio analysis is most useful when you are trying to explain something you can hear, or confirm something you cannot see from the filename alone. A waveform can reveal clipped peaks, long quiet sections, or uneven levels, while spectrum and frequency views help separate low-frequency rumble from hiss, hum, or overly bright content. Duration, channel layout, sample rate, codec information, and embedded metadata are also worth checking before you edit or convert a file; they often explain why two recordings that sound similar behave differently in another app.

Read measurements in context rather than treating one number as a pass or fail. A high peak does not automatically mean a file is too loud, stereo channels are not supposed to be identical in every recording, and metadata may be missing simply because an exporter stripped it. For troubleshooting, compare the suspicious file with a known-good recording from the same device or workflow. That usually gives you a better baseline than comparing against an arbitrary target.

Practical tips

• Check peaks and the waveform together; a single peak value says little about how often clipping occurs.
• Use a spectrum view to investigate hum, hiss, or rumble, but confirm the problem by listening before filtering anything.
• Compare left and right channels when a recording feels unbalanced or unexpectedly narrow.
• Record the original codec, sample rate, and channel layout before conversion so you can trace quality changes later.
• Treat tags and metadata as descriptive information, not proof of how the audio was actually encoded.