Computational parameters of the adjusted level (n, A-G characteristics)

Professional Definition

n – the number of frequency bands, – the value x of the frequency characteristic A, B, C, D, or G in the i-th frequency band.

The above description defines the variables necessary in the algorithm for calculating global sound level based on spectral data. To obtain the resulting level (e.g. $L_{Aeq}$), the acoustic energy from all n bands (usually octave or third-octave) must be summed. Before summing, an appropriate correction (gain or attenuation) value from the chosen weighting curve (A, C, G, etc.) is added to the level in each $i$-th band, which allows for obtaining the corrected result.

Acoustics in Simple Words

Imagine that acoustics is like cooking, and the result in decibels is the taste of the finished dish. The variable n is the number of ingredients you throw into the pot (e.g., bass, mid-tones, sopranos). However, before you mix them, you need to use a sieve – this is the value of the frequency characteristic. Some ingredients (like low rumbling) are heavily filtered through this sieve, while others (like human speech) come through entirely. Only after such selection (correction) of all n ingredients do we get the final "taste" of the noise, which corresponds to what the human ear perceives.

Summary

These variables form the mathematical basis for the operation of every sound level meter. They allow us to transform a complex frequency graph into a single, meaningful number that takes into account the specifics of human perception or the requirements of a specific industrial standard.