Multimedia File Manipulation
Image files versus audio files
When working with image files, there are a number of factors that influence the quality and size of the file. Many of these
have anologous factors when working with audio files. The usual goals of this lab are as follows:
- Understand how easy it is to manipulate an image,
- Understand the tradeoffs involved in image representation (file size vs quality for example), and
- Gain some experience with image manipulation software .
Instead of working with images, you are going to work with audio. However, given the intent of the lab, I do not want
you to spend your time learning the JAWS interface for a program you probably won't use again. If you already know iTunes
feel free to try out some of the examples given below, but otherwise just read through the comparisons. This lab will not have
a deliverable, and instead your lab and project grade will be combined (see Project description below).
Using iTunes to compare audio files
Consider trying to import music from a CD into iTunes.
There are a variety of import settings available from:
Edit > Preferences > Import Settings
One of the import options is to use WAV, which is a lossless compression format. By customizing the import options, we can compare to
some of the properties of images (you can change these in iTune by choosing "Setting"-"Custom"):
- Image Resolution is comparable to Sample Rate. In other words, the resolution (pixels per inch) you use to scan
a photo is comparable to the how often you sample the audio signal. Compare a higher (more frequent) sample rate vs. a lower
(less frequent) sample rate on the same CD track and see the difference in audio quality and file size.
(Make sure you use the same Bit Rate for each of these two samples.) The higher sample rate will result in better audio quality,
but at the expense of a larger file size.
- Bit (Color) Depth for images is comparable to Bit Rate for the audio sampling.
The former specifies how much precision you have in describing the quantities of Red, Green, and
Blue light for each pixel of the image. The latter specifies how many distinct levels of precision
you have for quantizing the continuous sound wave. Compare a higher (more precise) bit rate vs. a lower
(less precise) bit rate on the same CD track -- using the same, good quality rate sample rate for each --
and see the difference in audio quality and file size. Again, the more precise bit rate will result in a better quality
audio file, but a larger file size.
- Stereo vs. Mono is somewhat comparable to two layers of an image. Record the same track with the
same (good quality) sample rate and bit rate but make one stereo and one mono. It will be the same in
both "ears" in mono, different in stereo. But the mono should also be half the file size.
- To compare lossy compression schemes, you could also do a good quality sample with MP3
using the same settings as above and then compare. MP3 is somewhat comparable to the lossy compression of JPEG,
with lossless & uncompressed WAV more comparable to lossless & uncompressed version of TIFF.
So, the difference in audio quality should be noticeable unless the bit rate and sample rates are very high.
And, of course, the MP3 file should be considerably smaller than the WAV file. However, just as the lossy compression
of JPEG eventually becomes noticeable when you push the compression -- i.e., you get "compression artifacts" -- it's also
true that the sound quality of MP3 becomes noticeable as well, either from repeatedly exporting
(and re-compressing an already compressed audio file) and/or when the bit rate and/or sample rate are lower.
- For voice recordings, one can get by with lower bit rates and sample rates vs. musical recordings. Why might this be?
Project
On your machine in the lab (where JAWS is installed) I have had an audio editor installed.
The program is called Audacity, and is JAWS compatible. This is an
open source (i.e. free) tool for editing audio files (so if you just love it, you can install it on
your home machine as well!). For your project, I want you to spend up to 3 hours working on manipulating an audio file.
Turn in both the original audio file, and the final result of your manipulations. There are many audio files available for free online,
or else I can send you one to start (I suggest working with WAV files for the reasons described above).
You may find the following guide helpful in learning some of the JAWS key bindings:
http://vip.chowo.co.uk/wp-content/uploads/jaws/Audacity-1.3.11-Guide.html
Please note the 3 hour time limit!!! This is not intended to take your entire weekend, but instead give you a hint of what is possible.