In the early days of audio recording, sound was recorded as an analog signal. An analog signal is a continuous electric impulse of varying amplitude. Stored as vibrations on wax or plastic materials, the signal would be picked up by an electrically charged phonograph needle as the hard surface spun. The audio signal had a one to one correlation that allowed a precise reproduction. When technology shifted from analog signaling to digital signals, the format for recording signal changed from a linear model to a mathematically discrete pulse code moderation or PCM.
PCM breaks down the continuous linear signal into “points” called samples. If I were to compare a sound sample, it would be similar to a geological core sample for oil. Sound samples are mathematically calculated at certain points along the analog signal. The sample takes a measurement of the sound and converts the sound into a binary string. This binary string is inputted into the computer digital system as a raw data file.
The PCM’s audio stream maintains its fidelity by using two properties, sampling rate and bit depth. Sampling rate is the number of samples taken each second. The bit depth is the number of bits in a waveform. The size of a sampling rate is twice the frequency which for human hearing is between 20 Hz and 20,000 Hz or 20 kHz. So the sampling rate at it’s highest for human hearing is 40 kilohertz. Any higher and the sound pressure will damage hearing.