home | O'Reilly's CD bookshelfs | FreeBSD | Linux | Cisco | Cisco Exam  

Book HomeWeb Design in a NutshellSearch this book

24.3. Preparing Your Own Audio

Recording and producing your own audio requires a significant investment in hardware, software, and time spent learning. If you need to put professional-quality audio on your site but aren't likely to make the investment in time and equipment yourself, consider outsourcing the work to professionals.

The final product may be anything from a simple personal greeting to a live concert broadcast. The preparation of original audio requires a number of standard steps: recording, basic sound editing, then optimization for web delivery.

24.3.2. Basic Sound Editing and Effects

Once you've recorded raw audio, the next step is to clean up the recording. This can involve removing unwanted sounds, setting the beginning and end of the file, and/or making a loop. You may want to apply digital effects to the sound, such as reverb or a delay.

Consider also using mastering processing techniques such as normalization or compression that can balance out the level of your audio such that no part is too loud or too quiet.

There is a huge selection of software for audio editing and format conversion. The software ranges from single-purpose utilities available via free download to professional digital-audio editing suites costing thousands of dollars. Some popular professional-level tools are listed in the following sections. Windows audio tools

The following tools are available for use on Windows:

Sound Forge by Sonic Foundry


Sound Forge is limited to editing stereo files, but it includes many plug-ins for effects such as chorus, delay, distortion, reverb, and compression. Street price is about $400.

Cool Edit by Syntrillium

http://www.syntrillium.comFor only $69, Cool Edit offers up to four tracks of audio editing and includes plug-ins for audio restoration as well as the usual reverbs, delays and so on. At $400, Syntrillium's flagship product is Cool Edit Pro, which offers up to 64 channels in a full multitrack recording environment. Mac audio tools

These tools can be used on Mac systems:

Peak by Bias


With built-in batch processing and a street price of less than $400, this application has been the Mac standard when it comes to stereo editing. Bias also offers the more streamlined Peak LE for $99. This "Light Edition" may be sufficient for most entry-level users.

Spark by TCWorks

http://www.tcworks.de A more recent entry into the Mac stereo editing game, Spark is quickly making an impact. This program focuses on effects and offers support for VST plug-ins, but it also offers all of the editing features standard for professional work. Street price is around $400.

Digital Performer by MOTU


Performer software has evolved from a MIDI sequencing application into a full fledged digital recording studio environment offering top quality effects plug-ins and audio editing. Musicians can make entire recordings with this software, but it is also just as capable at adding audio to video or mixing radio programs. Street price is around $550. Tools for both Mac and Windows systems

These tools are available for both Windows and Mac systems:

Cubase by Steinberg


Similar to Digital Performer, this cross-platform multitrack recording environment offers both MIDI and audio editing with lots of effects plug-ins, virtual instruments, and recording tools for creating an entire virtual studio inside your computer. Street price is around $600.

ProTools by Digidesign


Long the industry standard for multitrack computer recording, ProTools offers everything you'd ever need for a professional quality recording studio in your computer. Their high end "Mix" systems, including both software and custom hardware, start at $7000 and go up from there, but they have recently started making consumer-level solutions such as the Digi001, which offers ProTools software and a hardware input/output box for around $900. Digidesign also offers "ProTools Free," a free version of ProTools (limited to eight tracks), at http://www.digidesign.com/ptfree/.

24.3.3. Optimizing for the Web

After the sound files have been recorded and edited, it is time to convert them to their target web audio format and make them as small as possible for web delivery. The tool you use may depend on the file format. For instance, RealAudio and LiquidAudio have their own creation tools. There are also several tools specialized for the creation of MP3s. Tools are discussed with their respective file formats later in this chapter.

One great all-purpose tool is Cleaner 5, from Terran Interactive available for the Mac and Windows systems. This program is designed to get the best quality files at the smallest size in whatever format you choose. Cleaner can compress a number of file formats, including Quicktime and RealMedia. It can also do batch processing. The program sells for $599 as of this writing. (Cleaner is the newer and renamed version of MediaCleanerPro.)

Regardless of the tool you use, there are standard ways to reduce the size of an audio file so it is appropriate for downloading via a web page. Not surprisingly, this usually requires sacrificing quality. The aspects of the audio file you can control are:

Length of the audio clip

It might seem obvious, but you should keep the audio sample as short as possible. For example, consider providing just part of a song rather than the whole thing. If you are recording a greeting, make it short and sweet.

Number of channels

A mono audio file requires half the disk space of a stereo file and and may be adequate for some audio uses.

Bit depth

Audio files for the Web are often saved at 8 bits, which will result in a file that is half the size of a 16-bit file. MP3s can handle 16-bit due to their efficient compression.

Sampling rate

Cutting the sampling rate in half will cut the file size in half (e.g., a sampling rate of 22.05 KHz requires half the data than one of 44.1 KHz). As a general guideline, audio files that are voice-only can be reduced down to 8 KHz. Sound effects work at 8 Khz or 11.025 KHz. Music sounds acceptable at 22 Khz.

Using these guidelines, if we start with a one-minute music sample at CD quality (10 MB) and change it to a mono, 8-bit, 22 Khz WAV file, its size is reduced to 1.25 MB, which is much more reasonable for downloading. Using MP3 compression, we can keep the quality of that one-minute sample at 16-bit, 44.1 kHz stereo (similar to CD quality) with a resulting file size of under 1MB. Combining these methods (a mono, 8-bit, 22Khz MP3), you can offer one minute clips at acceptable audio quality at only a few hundred K.

Obviously, just how stingy you can be with your settings while retaining acceptable quality depends on the individual audio file. You should certainly do some testing to see how small you can make the file without sacrificing essential audio detail.

Library Navigation Links

Copyright © 2002 O'Reilly & Associates. All rights reserved.