By Craig Anderton
Remember the dot-com boom? The web was supposed to be the force that democratized the music industry—physical distribution would be obsolete, as you could post your music on the web and be exposed to an audience of millions, who would cheerfully download high-quality music while paying a reasonable, and fair, fee.
Amazingly, significant parts of that have come true, and we’re not just talking iTunes: Bands set up web sites where fans can hear new tunes, MP3s make it easy to post songs for critiques, and snippets of music help sell physical CDs.
It’s not hard to get your music on the web, but unless you’re Trent Reznor, you’ll probably have to settle for posting it in a data-compressed format in order to conserve server space, as well as save time for those downloading your masterpiece. Although there’s a lot of criticism of the MP3 format, not all of it is justified as there are ways around some of its limitations.
The MP3 format is based on using data compression algorithms that can reduce the amount of data needed to reproduce music. (Note that this has nothing to do with audio dynamic range compression, as used in recording.) Actually these are data omission algorithms, because they do not work like StuffIt or Zip data compression algorithms, which restore the original file when uncompressed. Instead, a process like MP3 throws away “unneeded” data. For example, if there’s a lot of high-level sound going on, the algorithm might assume you can’t hear lower-level material, and decide for those sections you only need 24dB of dynamic range. This requires only 4 bits of resolution—25\\\% the amount of data required by 16-bit resolution.
Unfortunately, it’s difficult to retain quality with data-compressed music (video and images are much more easily compressed). One workaround is to use a lossless algorithm, such as FLAC, or the lossless options offered by Microsoft and Apple for their audio formats. However, these don’t result in particularly svelte files; with complex music, the size reduction may only be 10-20\\\%.
Although there are many data compression algorithms for audio, only a few are common:
When creating audio content for posting, even though MP3 doesn’t necessarily provide the best quality, all the players read it, there’s a ton of supporting software, and people can load the files into portable MP3 players.
When encoding a file to MP3, always give the encoder high quality, uncompressed material so it can make the best decisions on how to apply the compression algorithm. Then, choose the compression parameter values carefully. When saving to MP3, you can typically choose from a range of bit rates (number of bits that get transferred in a second), from 320 kbps stereo (excellent quality, but largest resulting file size) down to 8 kbps mono (good enough for dictation). Compressing a standard 28MB WAV audio file to 320kbps stereo MP3 results in a 6.4MB file; compressing to 8kbps mono yields a 0.16 MB file—a data reduction ratio of 175:1.
In addition to fixed rates, there are variable bit rate (VBR) options that dynamically optimize the bit stream according to the material being played back. This is not as universally compatible, so it’s usually preferable to use constant bit rates.
For best results, save a file using a variety of bit rates and sampling frequencies, in mono and stereo, and determine which combines best sound with smallest file size. Note that mono will usually have higher fidelity than stereo for a given file size. For example, with an MP3 128kbps file, the mono version “spends” that bandwidth on a single, high-quality file. Stereo generates two 64kbps streams—one for each channel—and 64kbps doesn’t sound as good as 128kbps. However, this means you give up stereo, which you probably don’t want to do.
Although there are plenty of ways to convert files to the MP3 format, iTunes is free, readily available, and works for both Mac and Windows. If you don’t have iTunes already, download it from www.apple.com and follow the instructions for installation. Then, convert using the following procedure.
1. Open iTunes, then drag the files you want to convert into the main iTunes window.
2. From the menu bar, go iTunes > Preferences, then click on the General tab (Fig. 1).
Fig. 1: Click on the General tab in iTunes to get started, then click on Import Settings to set up the MP3 file format characteristics.
3. Click on the Import Settings button (Fig. 2).
Fig. 2: The Import Settings dialog box is where you can choose the file format, and custom settings for that format.
4. With the Import Using pop-up menu, choose MP3. The other choices are relevant only if you’re ripping from a CD.
5. Choose the MP3 settings from the Settings pop-up menu. If you want to keep things simple, you can choose one of the default data rate settings of 128kbps, 160kbps, or 192kbps. The higher the data rate, the better the fidelity. But you can also choose Custom, which gives you multiple options (Fig. 3).
Fig. 3: Customize the quality and size of your MP3 file with the Custom Settings dialog box.
Now that the encoding parameters are setup, let’s encode your file. In the main iTunes window, right-click (ctrl-click) on the name of the track you want to convert, and select Create MP3 Version. Wait a few seconds for the conversion, and iTunes will create an MP3 copy of your file.
As there’s no obvious indication of file format in iTunes, if in the future you aren’t sure which is the original and which is the MP3 copy, right-click (ctrl-click) on the name and select Get Info (Fig. 4). You’ll see the format, sample rate, bit rate, channels, and other info. And that's all there is to it!
Fig. 4: Find out a file's characteristics with the Get Info option in iTunes.
Craig Anderton is Editor Emeritus of Harmony Central. He has played on, mixed, or produced over 20 major label releases (as well as mastered over a hundred tracks for various musicians), and written over a thousand articles for magazines like Guitar Player, Keyboard, Sound on Sound (UK), and Sound + Recording (Germany). He has also lectured on technology and the arts in 38 states, 10 countries, and three languages.