What does the sample rate and sample size of audio means?

Question

In a MP3 audio file, I know theres a bitrate of audio which tells the number of bits in seconds are being used by the player, in other words a measure of quality of audio.

But I'm confused with the terms sample rate and sample size. Are they independent of the bitrate or quality of sound? or Can it be explained in some understandable terms?

Solution 1

Here is a great article on all three terms you're asking about. To summarize, here are the three definitions:

1. bitrate: simply amount of data per second, this can vary within a file (variable bit rate) or have a static value
2. sample rate: a frequency of how many times audio is measured per second, usually measured in kilohertz (kHz); a usual number you might see is 44.1kHz. This is tied directly to bit depth or the number of bits measured in each cycle

So at this point, we have to do some math and see that bit rate is bits per second (usually measured in Mbits/seconds), so bit rate = sample rate x bit depth. As far as I know, your sample size would simply be one of these one-second chunks of data.

Doing the pure math will reveal that these files would be very large, however there are a number of compression algorithms employed to keep the files lower without losing significant quality.

Solution 2

Sample Size or Bit Depth is a measure of how many bits a sample contains which is directly a measure of quality. But this applies only to PCM Sampling, For lossy formats like mp3, Sample size doesn't really define the quality.

Please visit Audio Bit Depth to learn more.

Solution 3

Sample rate= Sample rate is no. of samples of audio carried per second

Sample size= The sample size is determine the maximum dynamic range of a digitized sound. Dynamic range is the ratio between the highest amplitude and the lowest non-zero amplitude in a signals and usually expressed in decibels (dB)

Comments

[Ƭᴇcʜιᴇ007]

Have you seen this yet?

[sleske]

Note that for uncompressed audio, file size is indeed "sample_rate * bit_depth * no_of_channels * audio_size_in_s". Hence an hour of CD quality audio has 44,100*16*2*3600 =5,080,320,000 bit , or about 606 MiB.

[jayarjo]

So how would you use this information - sample size and fidelity of the compression?