Intensity

Intensity is a measure of how loud or strong a signal is. VoceVista shows it in two complementary ways: the overall intensity of the signal at each moment, and the intensity of each frequency component in the signal.

Overall and Per-Frequency Intensity

The Waveform shows the combined strength of all frequencies at each instant in time. Values in the waveform are amplitudes, scaled into the range -1 to +1 for digital audio. When the signal exceeds those limits, samples are cut off at ±1, producing distortion known as clipping.

The Spectrum and Spectrogram show how the energy is distributed across frequencies. The Spectrum shows intensity as height on its vertical axis; the Spectrogram shows intensity as colour. Reading these views together — overall intensity in the waveform, frequency-by-frequency intensity in the spectrum — gives a complete picture of the loudness structure of a recording.

Decibels

Spectrum and Spectrogram measure intensity in decibels (dB). The dB scale is logarithmic and relative to a reference level, and is built around three rules that take some getting used to:

  • 0 dB is the maximum — the loudest signal the file format can represent. There is no positive dB value; anything stronger would clip.

  • All other values are negative: -6 dB, -12 dB, -24 dB, and so on.

  • Each -6 dB step is roughly a halving of perceived loudness.

So -6 dB is half as loud as 0 dB, -12 dB is a quarter, -18 dB is an eighth, and so on. The scale extends downward indefinitely; the practical floor is set by the noise of the recording and by the bit depth of the file.

Why the Scale Is Inverted

This arrangement seems backwards at first, but it reflects a real asymmetry of digital audio:

  • There is no natural lower limit to how quiet a sound can be. You can always make a vibration smaller; the bottom of the scale runs to negative infinity.

  • There is a natural upper limit: the largest number the file format can encode. That hard ceiling becomes the obvious reference, and so it is set to 0 dB.

A useful analogy is depth below the surface of a body of water. The surface is the natural reference (0 m); every depth below it is negative (-1 m, -2 m, …); and there is no theoretical limit to how deep one can go. Saying "I’m at -10 m" gives a clear, absolute position. Saying "I’m at 50 % of the depth" doesn’t, because the answer depends on how deep the water is at that point. Sound intensity in dB works the same way: 0 dB is the surface, and every quieter signal sits somewhere below it.

Dynamic Range

The actual range of intensities you can see depends on the recording format and on the analysis:

Recording format

16-bit audio supports about 96 dB of dynamic range; 24-bit, about 144 dB; 32-bit float, an unrealistically wide theoretical range — useful primarily as headroom against clipping during recording and processing.

Spectrum analysis

The spectrum display typically shows usable detail across about 100 dB. The bottom of that range is set by the FFT window function, by background noise in the recording, and by the microphone.

For most analytical purposes, focusing on the range from 0 dB down to -60 dB shows the relevant content. Sounds quieter than -60 dB are usually indistinguishable from noise.

In VoceVista

The visible dynamic range of the Spectrum and Spectrogram is set with the brightness and contrast controls on the toolbar; the Waveform’s amplitude axis can be switched between linear and logarithmic display modes, see Timeline and Waveform. Colour mapping for the Spectrogram is configured in the Colormap Editor.