|
Human ear
A key concept in all methods of encoding audio digitally is how the human auditory system or ear actually perceive sound. The most common feature is the idea of masking. All of us have an absolute threshold of hearing (ATH) that varies with frequency, age, gender etc. The ATH is the minimum sound pressure, level or amplitude of a single pure tone that the ear can hear in a noiseless environment. If a sound is below that curve then we just won’t hear it.
Masking is the concept that certain sounds render other sounds inaudible. Masking is either simultaneous where both the wanted and unwanted sounds occur at the same time or non-simultaneous where the unwanted sound can occur either before or after the wanted one.
Masking effects are frequency dependent and so the audio spectrum can be split down into sub-bands and hence a set of sound masking curves.
Finally we come to the idea of a Masking Threshold. Consider the effects of a high level 1 KHz tone appearing suddenly in an audio stream. Say the 1 KHz tone affects the human ear from 800 Hz to 1500 Hz. We can now draw a curve between these points at a certain sound pressure level and be confident that any sounds below this masking threshold will be inaudible to the human ear.
|
