9 Replies - 42976 Views - Last Post: 20 August 2013 - 10:49 AM

#1 Sinned  Icon User is offline

  • D.I.C Head

Reputation: 18
  • View blog
  • Posts: 207
  • Joined: 13-October 10

Deep understanding of digital sound

Posted 16 September 2012 - 05:51 AM

Hello everyone,

I realized that sound is just made of some (analog) waves, running on different frequencies and strengths.
So I went looking on the internet about how this works and how this could be saved on a computer.

WAV files appears to be the most accurate and "simple" music files. (uncompressed)
But now I still don't know what these files actually are, what they contain and how they can be used to produce sound.

I don't really know how to ask this questions, but I think this would make the point:
Does anyone know a documentation which explains the way sound works and how this is digitized (and vice-versa)?

Thanks in advance,
Sinned

Is This A Good Question/Topic? 0
  • +

Replies To: Deep understanding of digital sound

#2 cgren72  Icon User is offline

  • New D.I.C Head

Reputation: 3
  • View blog
  • Posts: 35
  • Joined: 29-August 12

Re: Deep understanding of digital sound

Posted 25 September 2012 - 08:50 AM

Can you call a sound waveform "analog" or "digital?"
I am not really sure of this, so I shouldn't make a fool of myself, but when an audio file is rendered to wav, I think it is recorded and turned into computer readable data and stored in wav format. And when it is played, it is opened and read.
Was This Post Helpful? 0
  • +
  • -

#3 Sinned  Icon User is offline

  • D.I.C Head

Reputation: 18
  • View blog
  • Posts: 207
  • Joined: 13-October 10

Re: Deep understanding of digital sound

Posted 26 September 2012 - 06:39 AM

As far as I know "computer readable data" are binary numbers. (0 and 1)
A digital signal is a block signal, from which the up's are seen as 1s and the down's as 0s.
So a digital signal can easily put on a computer and read from it

My question is actually about how this digital (block) signal is converted into the analog "waves", because I don't understand how those bits could get converted into a waving signal from various frequencies and strengths.
Was This Post Helpful? 0
  • +
  • -

#4 cgren72  Icon User is offline

  • New D.I.C Head

Reputation: 3
  • View blog
  • Posts: 35
  • Joined: 29-August 12

Re: Deep understanding of digital sound

Posted 26 September 2012 - 08:14 AM

Yeah thats what I meant, I just didn't want to say ones and zeros, on off, up down, or whatever. Does wav use an algo like mp3?
Was This Post Helpful? 0
  • +
  • -

#5 Sinned  Icon User is offline

  • D.I.C Head

Reputation: 18
  • View blog
  • Posts: 207
  • Joined: 13-October 10

Re: Deep understanding of digital sound

Posted 26 September 2012 - 02:09 PM

I don't think wav uses any sort of algorithm to compress the content.

However again I don't know HOW the analog signal gets converted into the digital one. (And vice-versa)
Was This Post Helpful? 0
  • +
  • -

#6 Utael  Icon User is offline

  • D.I.C Head

Reputation: 55
  • View blog
  • Posts: 210
  • Joined: 12-December 11

Re: Deep understanding of digital sound

Posted 27 September 2012 - 09:55 PM

It's actually not as simple as just 1s and 0s each frequency (depending on the bit rate, the higher the bit rate the more "resolution" the file has) is assigned 8 bits or more, again depending on the resolution or quality, for how much that frequency has in power for that section of time. Those add up from 20hz to 20khz to form the actual sound. Add sections together you make the sound waves. I'll try and find more info and links in the morning.
Was This Post Helpful? 1
  • +
  • -

#7 grg  Icon User is offline

  • New D.I.C Head

Reputation: 0
  • View blog
  • Posts: 1
  • Joined: 02-February 13

Re: Deep understanding of digital sound

Posted 02 February 2013 - 03:43 PM

View PostSinned, on 26 September 2012 - 02:09 PM, said:

I don't think wav uses any sort of algorithm to compress the content.


It depends how you define 'compress' acutally. Some adress it even to simple quantization.


Quote

However again I don't know HOW the analog signal gets converted into the digital one. (And vice-versa)


Check Pulse-Code Modulation http://en.wikipedia....tion#Modulation
Was This Post Helpful? 0
  • +
  • -

#8 BBeck  Icon User is offline

  • Here to help.
  • member icon


Reputation: 570
  • View blog
  • Posts: 1,270
  • Joined: 24-April 12

Re: Deep understanding of digital sound

Posted 19 August 2013 - 10:11 PM

Basically a .wav file is two sound waves. One for the left channel (speaker) and one for the right channel (speaker).

Very simply the digital wave is converted into an electrical wave that is for all practical purposes identical. This wave pushes the speaker cone in your speaker in and out and creates an actual sound wave that is also identical to the digital wave for all practical purposes.

Recording reverses the process. If you want a low fi recording you could actually sing into your speaker, produce an electrical wave, and record that electrical wave to a .wav file digitally. Speakers, not surprisingly, are not designed to record sound; it will work, just don't think you're going to record your next album this way and have people say it sounds good.

Back in the old days, with Thomas Edison and those guys they recorded to acetate (EDIT: Actually Edison recorded on tin and then wax Amberols. Acetate was used starting about the 1930s). Eventually, they started recording to vinyl. Basically, the microphone vibrated as you sing into it, the vibrations of the diaphragm in the microphone converted this into an electronic Alternating Current (AC) wave which then would cause a needle to vibrate and cut groves into the spinning vinyl disk.

Rotating the disk while in contract with a playback needle would cause the needle to vibrate and turn the waves recorded on disk into electronic waves, which could then drive a speaker to turn the electronic wave into a physical sound wave.

In fact, in the oldest of days it wasn't even electronic. It was just a speaker, a needle, and a "disk". No electronics what so ever. But the sound wave was physically cut into the disk by the vibration of the cutting needle. A recording needle could play it back without damaging the disk.

So, fast forward a billion years or so into modern times and we have the digital world with .wav files.

Basically the same sound waves are recorded into the file as went onto those disks way back when. Once the wave is turned into an Alternating Current (AC) electronic wave by the microphone, it can then be turned into a digital wave by an Analog to Digital converter.

The AD converter takes a sample of the wave every so often. For CDs it was determined that it should take a sample 44,100 times per second. This is because humans generally cannot hear a sound that vibrates faster than 20,000 times per second. Doubling that allows for good sound resolution (it sounds better probably because if it where 20,000 you might actually hear slight imperfections, but the bottom line is that they felt most people would not be able to tell the difference if they increased the sampling speed higher than 44.1k.


So, for CDs they sample the sound wave 44,100 times per second and they measure the sound wave's height at that point in time. This height is stored digitally by a certain number of bits. So it might be 8 bits, 16 bits, 24 bits. The more bits you have storing wave height, the more accurate the reproduction of your wave is going to be. But the more bits you use the larger your data file is going to be.

So, by breaking the sound wave up into 44,100 time slices per second and recording the height of the sound wave during that slice and converting it into bits, you get digital sound of "CD quality" which is typically what goes into a .wav file.

In the studio, you record basically two wave channels/tracks to produce a "stereo image". In other words, there is a separate channel/track/sound-wave for the left speaker and another for the right speaker.

This post has been edited by BBeck: 20 August 2013 - 01:45 PM

Was This Post Helpful? 1
  • +
  • -

#9 BlueMelon  Icon User is offline

  • D.I.C Head

Reputation: 40
  • View blog
  • Posts: 187
  • Joined: 27-April 10

Re: Deep understanding of digital sound

Posted 20 August 2013 - 10:38 AM

View PostBBeck, on 20 August 2013 - 05:11 AM, said:

The AD converter takes a sample of the wave every so often. For CDs it was determined that it should take a sample 44,100 times per second. This is because humans generally cannot hear a sound that vibrates faster than 20,000 times per second. Doubling that allows for good sound resolution (it sounds better probably because if it where 20,000 you might actually hear slight imperfections, but the bottom line is that they felt most people would not be able to tell the difference if they increased the sampling speed higher than 44.1k.


http://en.wikipedia....ampling_theorem
Was This Post Helpful? 0
  • +
  • -

#10 BBeck  Icon User is offline

  • Here to help.
  • member icon


Reputation: 570
  • View blog
  • Posts: 1,270
  • Joined: 24-April 12

Re: Deep understanding of digital sound

Posted 20 August 2013 - 10:49 AM

I prefer this explanation for non-electrical engineering majors. :-)

http://electronics.h...og-digital3.htm

Wave files are actually pretty simple. It's really just a digital version of the actual sound wave that "hit the microphone". (In reality, it was probably a sound wave that was created on a multi-channel mixing board across numerous "takes" with a few dozen microphones or possibly some of the instruments plugged directly into the console or even only existing in the digital space such as sampled instruments. And all of that is combined into the left and right channels of the stereo wave file that goes onto the CD. MP3 and other formats have compression and things that make them more complicated but there is no compression in wave files; it's basically just a digital copy of the actual sound wave that "hit the microphone".)
Was This Post Helpful? 0
  • +
  • -

Page 1 of 1