Next: Typical Audio Formats
Up: Basics of Digital Audio
Previous: Nyquist's Sampling Theorem
Affects Quality of Audio
 Ears do not respond to sound in a linear fashion
 Decibel (dB) a logarithmic measurement of sound
 16Bit has a signaltonoise ratio of 98 dB  virtually inaudible
 8bit has a signaltonoise ratio of 50 dB
 Therefore, 8bit is roughly 8 times as noisy
 6 dB increment is twice as loud
Signal to Noise Ratio (SNR)
 In any
analog system, some of the voltage is what you want to measure (
signal), and some of it is random fluctuations (noise).
 Ratio of the power of the two is called the signal to
noise ratio (SNR).
SNR is a measure of the quality of the signal.
 SNR is usually measured in decibels (dB).

Typically 8 bits or 16 bits.
 Each bit adds about 6 dB of resolution, so 16 bits => 96 dB.
 Samples are typically stored as raw numbers (linear format
), or as logarithms (ulaw (or Alaw in Europe)).
 Logarithmic representation approximates perceptual uniformity.
Affects Size of Data
There is therfore is a trade off between Audio Quality vs. Data Rate
Some typical applications of sample bit size and sample rate are listed below:
Quality Sample Rate Bits per Mono/ Data Rate Frequency
(KHz) Sample Stereo (Uncompressed) Band
     
Telephone 8 8 Mono 8 KBytes/sec 2003,400 Hz
AM Radio 11.025 8 Mono 11.0 KBytes/sec
FM Radio 22.050 16 Stereo 88.2 KBytes/sec
CD 44.1 16 Stereo 176.4 KBytes/sec 2020,000 Hz
DAT 48 16 Stereo 192.0 KBytes/sec 2020,000 Hz
AUDIO DEMO: Comparison of Sample Rate and Bit Size
Click on the file links below to audibly hear the difference in the sampling rates/bit sizes indicated for each file type:
 Telephone uses ulaw encoding, others use
linear. So the dynamic range of digital telephone signals is effectively
13 bits rather than 8 bits.
 CD quality stereo sound > 10.6 MB / min.
Practical Implications of Nyquist Sampling Theory
We'll finish off with a question: Why are CD Sample Rates 44.1 KHz?
The answer should be obvious if you have paid attention to the above notes (Answer in Lecture).
Next: Typical Audio Formats
Up: Basics of Digital Audio
Previous: Nyquist's Sampling Theorem
Dave Marshall
10/4/2001