(compMus1) Digital Audio Properties

Digital audio quality is determined by two properties: sampling rate and bit resolution.

The sampling rate determines how often amplitude has been measured per second (per channel). The sampling rate determines the highest frequency that can be represented accurately. To get an accurate measurement for a particular frequency, you must measure amplitude at least twice per wave cycle. The result is that the highest frequency you can represent accurately is half the sampling rate. This half-point frequency is called the Nyquist Frequency. For a CD-quality system, the sampling rate is 44,100 Hz, and the Nyquist Frequency is 22,050 Hz. Keep in mind that halving the sampling rate will only reduce the pitch range by one octave. High Definition sampling rates provide better time resolution, which can help in localizing audio.

The bit resolution determines the accuracy of the amplitude measurement. Inaccurate amplitude measurements result in audible noise, called quantization noise. The greater the error, the more audible the noise. Each bit of resolution doubles the accuracy of the amplitude measurement, resulting in a 6 dB increase in the signal-to-noise ratio. CD-quality audio is recorded at 16 bit resolution. Higher bit resolutions provide less noise.

Bits is short for binary digits. Binary counting is base two counting, meaning that there are only two digits per place (a zero or one). In any counting system, each place is the base to the xth power, starting with zero power. So, the binary number 1011 can be translated to base ten as:

1 x 2(0 power) = 1
+ 1 x 2(1 power) = 2
+ 0 x 2(2 power) = 0
+ 1 x 2 (3 power) = 8
= 8 + 2 + 1 = 11


Leave a Reply