When storing images the picture is separated into pixels, and the pixels’ colors can be defined with the RGB scale. So the computer can digitally store and send images as numbers. How does it do the same with sound files?
In: Technology
These answers seem to be forgetting the ‘li5’ part. Let me start by explaining how sound is recorded. Sound vibrates a magnet in a microphone. That vibration is transferred into voltage variations(vv). This all happens in your mic. The vvs travel down the mic cable and from there is ran through a box that encrypts the intensity of the vvs as good ‘ol 1s and 0s. From there, it can be stored as data in a computer. You basically use the same process in reverse to go through the speakers. That’s why there are magnets in speakers. Hope this helps [:
(This is fairly generalized and there are different variables, but to sum it up, the computer doesn’t see music, it sees voltage variations.)
Simplest way to understand is picture a typical sine wave, with a horizontal line through the middle of the wave. Now pick a point above the line and draw a line from this point to the middle line (and its to be perpendicular)
Computers store sound as a series of these measurements. A positive number says the point is above linr, a negative number says the point is below the line.
You’ll see sound quality represented as both a frequency, and a bit value eg 8bit, 16bit, 24bit.
Frequency represents how many of these measurements you’re making per second. So 44Khz sound you have 440000 measurements to represent the sound.
The bit value is the resolution of your measurement. So for 8bit sound, the maximum value/range/resolution you can measure is +127, the lowest is -127 and it’s one byte per measurement . For 16bit it’s ±16384, and two bytes per measurement. For 24bit sound it’s 3 bytes per measurement.
Now if the sound is stereo then that’s two streams of sound, one channel for left and one for right. Each stream is stored separately. If the audio is Dolby 5.1 then that’s 6 audio streams, all stored separately.
So putting it all together, if you have stereo 16bit 44khz audio, that’s 2×2×44000 bytes of data to store one second of sound.
Now there are various compression and encoding algorithms to store sound data so it takes less bytes, but this is the simplest way to store sound digitally.
As far as I understood it is pretty similar to what you described with an image. With images you have the RGB to display all your colors. In sound these are all various frequencies. And instead of aligning them in 2D what you’d do in a picture you align the frequencies in a series/time. That’s how I understood it but by all means no expert in this field
I guess, analogous to pixel data, you store combinations of frequencies and duration.
But that’s very simplistic, same as saying images are stored by pixel rgb values. There are tons of image formats and there are tons of sound formats. They all determine how the data is compressed and physically stored in the file.
Heck there are drastically different image formats, that don’t store pixel data, SVG for example. There is the same for audio too, MIDI.
It’s all the same thing. The RGB subpixels are communicated using a series of 1 and 0s that when put into a program (like the one in your GPU) it puts together the array of subpixels required). When your audio program saves audio, it changes it into 1 and 0s which when played using a music player, plays audio.
* Sound is captured digitally in a two step process:
* First sound is converted to a varying voltage through a microphone.
* The microphone has a small piece of material that vibrates when a sound is made.
* These vibrations cause a small device called a transducer to create a tiny voltage that mimics the movement of the air due to the sound.
* Then an electric device measures that voltage thousands of times a second and stores the value in digital form.
* Those digital values are now just like any other piece of digital data and can be stored or transmitted just like your emails or a video.
Latest Answers