I’ve always wondered how audio can be recorded to electricity then played back as perfect audio.
In: Technology
A microphone essentially measures very small air pressure differences (which is what sound is – rapid changes in air pressure). So the digital part comes in when the circuitry assigns a number to the pressure level say from 0 to 100 (in reality it is larger than this) and takes the pressure reading many times a second. This gives a long series of numbers which represent the pressure level at different times. This is digital information (just numbers) and can be stored as a file on a hard disk or compact disk or sent to the internet.
To play back the sound, it does this in reverse. The electronics reads the numbers and outputs an appropriate signal to the speaker and does this using the same time interval used during recording. The speaker converts the signal to movement of the driver which causes a pressure variation (ie sound)
The computer or electronics doesn’t “understand” sound. It simply converts a continuous signal (from the microphone) to a series of numbers and the numbers back to a signal for playing back.
Sound is made up of vibrations – things moving back and forth a little bit in regular patterns.
Imagine a beach. The waves go up and down in regular ways. If you sat on the porch and marked down the exact time each wave crashed onto the sand, you could reproduce the rhythm of the waves later. That’s essentially what an electronic recording does – it records the movement of a little reed (thin vibrating sheet) inside a microphone, then makes a reed inside a speaker repeat those movements later.
Digital audio is made up of a sequence of samples. Each sample is a loudness and if they are played back fast enough, you can recreate the recreate the original sound waves. In math, there is a theorem which describes how many samples you need to create waves: Nyquist–Shannon sampling theorem. In order to fully recreate the waves, you need to sample twice as fast as the highest frequency in the signal. Human hearing can detect sound waves from frequencies from about 20 Hz to 20 kHz, so if the digital audio samples at 40 kHz with enough precision in the loudness levels it can fully recreate the sound.
Now, this doesn’t mean that the digital audio is going to sound the same as playback from a non-digital audio device. Digital audio is almost exclusively done using electronics, while non-digital audio was done mechanically, and the mechanical reproduction usually acts as a low-pass filter, removing some of the high frequency noise. This is why the sound for a record player doesn’t sound the same as from an electronic speaker.
Latest Answers