At the end of this module you should be able to the following:
Describe the process by which sound is turned into bits
Describe factors that determine sound quality, including the sampling rate
Describe how sampling errors are produced
Discuss how compression can affect sound quality
Describe in general terms the process of perceptual coding
Name some popular file formats for digital audio
Name some tools for creating and editing digital audio
Just like text and images, sound can be represented and stored digitally. However, digitally representing and storing sound is slightly more complex than storing images or text, in part because sound is temporal (that is to say, it unfolds through time) and because of the near-infinite variety of sounds that we can hear ( compared to the limited number of characters in any one alphabet or writing system).
Allot 20 minutes
Read HowStuffWorks.com's Web article on digital and analog sound by clicking on the links in the list below to view the pages you should read.
An introduction to the material in this Web article
A representation of sound can be recorded onto a solid medium of some sort
A continuous sound wave can be broken down into discrete units called samples
An example computation of the amount of bytes needed to store a 74-minute long sound file.
After finishing the tutorial, answer the following questions for yourself:
Which has greater fidelity, an analog recording or a digital recording of a signal?
What considerations would you take into account when recording a speech or a lecture as opposed to recording live music?
In Activity 6.1, you may have noticed that the example of a 74-minute long sound file would need close to 747,000,000 bytes if you were to store it in its raw form. That's around three hundred seventy-four 2Gb USB flash drives! If you were interested in serving the sound file over the Web, that's still a lot of data to push over even in today's terms, so we often have to compress audio data.
Compression allows you to minimize the amount of data that it takes to represent digital audio. The most common type of compression uses perceptual coding, which makes an assumption about what average human beings cannot hear and eliminates it from the sound file. For example, very high-pitched and very low-pitched sounds could be discarded, as well as very soft sounds that follow right after very loud sounds. MP3 is an example of an audio format that uses perceptual coding.
Allot 15 minutes
Read this short article on the Telecom ABC website about perceptual coding.
Listen to the Susanne Vega song, Tom's Diner; a YouTube video of the song is embedded on this page for your convenience.
Finally, read this Gizmodo review of a project that retrieves the sounds that are typically discarded from the Tom's Diner song. Make sure to watch (and listen) to the video embedded on the page.
Can you hear the faint strains of Tom's Diner in the Ghost in the MP3 video? Do you agree that these sounds--when removed from the original song--are barely detectable to most individuals? Does this video give you a better appreciation for how perceptual coding in audio works?
Video 6.1. Susanne Vega's song, Tom's Diner
When saving a sound file after creating and editing it in your audio software, you will probably have the option of saving it in any of several file formats. Some popular file formats are WAVE (.wav), MP3 (.mp3), and Ogg Vorbis (.ogg). MP3 is by far the most popular format, but it is actually a proprietary format. Ogg is open source, but is not widely supported. You can learn more about audio file formats in a concise summary provided by an audio software company, NCH, on http://www.nch.com.au/acm/formats.html.
Because sound can be represented as waves that can be stored digitally as sequence of bits, you can not only change how an audio file sounds by manipulating the various features of the wave, but you can even create your own sounds. We will not cover sound manipulation and creation in this course, but you may learn more about it in the Audio in Multimedia course, MMS 172.
This module provided you a rough overview of the technical considerations related to recording, storing, and playing back digital audio.
While you may think that our sense of sight is more important than our sense of hearing, humans are actually very sensitive to sound. For example, studies have shown that when people are having a videoconference call over a poor connection, people are more tolerant of errors in the image quality than with errors in the sound. That is to say, people will often find dropped audio more displeasing and annoying that dropped visual information. In addition, recorded sound can have effects on people that analog or live sound may not be able to induce. You will learn about some unusual and powerful examples that illustrate the power of sound on people in Unit 3.
Brain, M. (n.d.). How Analog and Digital Recording Works. Retrieved January 19, 2019, from https://electronics.howstuffworks.com/analog-digital.htm
Dvorsky, G. (2015, February 17). Listen To All The Stuff MP3 Compression Leaves Out. Retrieved January 19, 2019, from https://io9.gizmodo.com/listen-to-all-the-stuff-mp3-compression-leaves-out-1686279060
kittykittymewomeow2. (2008). Susanne Vega - Tom’s Diner. Retrieved from https://www.youtube.com/watch?v=-26hsZqwveA
Maguire, Ryan. (2014). The Ghost in the MP3. Retrieved January 19, 2019, from http://theghostinthemp3.com/
NCH Software. (n.d.). File Formats. Retrieved January 19, 2019, from https://www.nch.com.au/acm/formats.html
Telecom ABC. (n.d.). Perceptual audio coding - Telecom ABC. Retrieved January 19, 2019, from http://www.telecomabc.com/p/perceptual.html