Sound Library Construction

Apr 1, 1999 12:00 PM, By Gino Robair

Sound libraries, such as sample CDs, are ubiquitous in the personal or commercial studio. If you have ever used them, you may have wondered how they’re put together. Perhaps you have amassed an archive of recordings or instrument-specific patches and have thought about compiling your own sample collection, if you only knew where to begin.

Armed with the items typically found in a personal studio (DAT machine, DAW, CD burner, sampler), you can create a professional-sounding library to meet your needs. All it takes is some vision, planning, inspiration, and a good deal of editing.

I surveyed a handful of successful professional sound designers, manufacturers, and distributors about the various aspects involved in creating a sound library. Whether you’re recording breakbeats, exotic instruments, production music, sound effects, or any combination thereof, the following information will help you maximize your time and energy as you organize your favorite sounds for personal use or commercial distribution.

From the Top Down

The first order of business in making a quality sound library is to do some pre-production planning. Begin by determining your goals for the project: Is it for personal use or will you be sharing it with others? What are the intended uses for these sounds? What format will work best? Do you plan to license it to distributors? Setting a goal for your project will help you answer many of the questions that will arise as you get further into it.

For example, your plan might be to create and license a library in a specific sampler format to an established company. In this case, you’ll need to have the most recent update of the machines that read the format you’ve chosen, with the maximum amount of RAM available. This will allow you to take full advantage of every nuance that machine has to offer.

Perhaps your objective is simply to create an audio CD for your own post-production sound-design purposes, in which case you’ll want to concentrate on your recording and CD-burning setup.

Remember that you can always modify any goals that you set for yourself as the project progresses. After all, you may find yourself heading down a new path once you’ve gotten started.

ORGANIZATION

Anyone who has used a sound library will tell you that a well-organized disc is worth the money it costs to create or buy it. Determining the best way to organize a collection of sounds depends on how the library will be used, as well as the taste and style of the user.

Sound designer Jeff Darby assembled the 25-disc sound-effects library for Earwax, a production company in San Francisco. The discs are housed in a large three-ring binder with a log that breaks down each disc into specific categories and subcategories.

“The Earwax library is made up of Sound Designer II files, primarily. Each CD-R is a data disc containing Mac files and folders, burned with Adaptec Toast,” he explains. “If you’re using Pro Tools, you import the files directly into the application. This method eliminates a couple of steps: you don’t have to edit, normalize, or digitize sounds each time. Once they’re done, they’re done.

“The Earwax library is mainly a byproduct of jobs we’ve done,” he continues. “We then organize the sounds from the jobs into categories such as Whooshes, Machines and Tools, and Weapons and Violence. In the Whooshes folder, you’ll find a folder of swishes and perhaps one of dopplers. Weapons and Violence contains subcategories such as Hits and Thuds. The Thud folder may, in turn, contain 45 different thuds. Usually, there are a lot of subcategories.”

Organizing sounds into categories is a subjective art and is often a matter of personal preference. “In the Ambience folder, for example, you might have nature, city, or bird ambiences; I wouldn’t necessarily put a bird ambience in an animal folder. There are a lot of gray areas. We usually build the library according to the way we think and what our needs are.”

Earwax has considered putting together a “best of” collection for commercial release. “At the rate it’s going,” says Darby, “we’ll probably have another ten discs in the library by the end of the year, which will give us plenty to choose from.” Their biggest challenge will be categorizing the material in such a way that it will be intuitive for a new user.

Sounds are organized in libraries in a number of ways. Here are some of the more common ones.

Instrument. If the project features a collection of instruments, grouping each instrument or class of instruments together makes good sense.

Key. Compiling sounds by key is an especially useful way to organize a library of instrument-specific riffs and licks (see Fig. 1).

Tempo. If the disc will be used in a dance or club environment, having the samples arranged by tempo in bpm (beats per minute) is essential. An elegantly designed library will first give you the entire groove (full orchestration) at a given tempo, followed by each component of the groove, individually.

Theme or genre. This does not necessarily mean a “musical” theme. If the library is a sound-design collection for post-production, organizing by theme (for example, sci-fi, western, and ethnic) would be a logical way of working. If you are assembling a library of music cues, you could group tracks by genre (classical, rock, jazz).

Effects type. This category involves classifying effects by how they sound (for example, pops, applause, and gunshots) or what their functions are (ambiences, hits, effects).

Combinations. For a comprehensive collection of samples, you may need to combine some of the above organizations. For example, you could divide one portion of an orchestral disc into instrumental sections (say, brass and woodwinds); another into orchestration (groups or soloists); and so on, on down to individual notes and effects. Another section of the disc could be organized into tempo- or key-related selections.

DOCUMENTATION

A user-friendly library relies on full documentation. A great-sounding disc becomes difficult to use if the liner notes are cryptic or nonexistent. If you are planning to license your library, thorough documentation is a must and begins the moment you press Record.

Doug Morton of Q Up Arts recommends keeping detailed notes throughout the recording session. “Make sure you document what you’re doing so that everything will be clear to the person who works on the material later on.”

Your recording notes will be useful as you edit and loop your samples. Notes on mic choices, mic position, recording location, and any processing that occurred, are handy when you need to combine different parts of a sample or create Velocity layers. Such details may seem mundane at the time, but they may save you from having to rerecord elements later.

Your session notes will also be helpful when you compile your liner notes. The kind of documentation you publish with your library will depend on the format you’re working in, as well as its intended use. Notes for an audio CD of loops and grooves should include the name and number of each track, tempo, number of bars in each groove, start times, track length, key, and instrumentation. When the groove is broken down into individual components, these should be noted as clearly as possible.

For a CD-ROM release, it is crucial to include format-specific information in your documentation. A library in the Akai format, for example, should give the location, volume, program name, sample size, and length of bars of each sample. In addition, you might include key, tempo, crossfading, and Velocity-switching information, as well as whether the sample is stereo or mono. Providing details about each instrument or voice that is sampled is also useful.

Spectrasonics’ Symphony of Voices CD-ROM is a four-disc set that comes with two elegantly produced, 50-page booklets. Booklet 1 is a CD-ROM directory that includes setup instructions, system recommendations, and a list of patches and performances. Every sample is listed with key and tempo notation. The liner-note booklet includes biographies of the artists, performance tips, an overview of the organization of the sounds, descriptions of the recording and editing equipment used, the philosophy behind the project, and a list of frequently asked questions about using the company’s sound libraries (see Fig. 2). Such a cornucopia of documentation enhances the users’ knowledge about the sounds as well as the pleasure they get from using the library.

BRAIN STORM

When coming up with ideas for projects, Eric Persing, creative director at Spectrasonics, begins by imagining what he would like to hear in a sample library. “I might begin by thinking about what’s not in our catalog, or by considering what people are interested in.” Product-registration cards give him a sense of what customers are looking for in new products.

“Other times it’s a whim,” he continues. “Sometimes we’ve started with just the title. Distorted Reality, our most popular release, began that way. We also have to consider what we are capable doing. Thinking things through is essential because [creating a sound library] is a long process. Several of our libraries have taken up to two years to complete.”

Morton says that at Q Up Arts, quality and uniqueness are what drive new product development. “We try to find something that’s different, or find something that hasn’t been done well and do it better. We also try to avoid stylistic trends. Our niche is world/ethnic/esoteric content, such as Voices of Native America. It’s the harder-to-find kinds of things that we prefer.”

FORMAT ROULETTE

Whether or not you’re interested in going commercial with your sounds, a brief discussion of how companies choose a format for their sound library is useful—especially if you’re planning to purchase any gear in the near future.

There’s a wide range of formats to consider, the most common of which are Akai, E-mu, Ensoniq, Kurzweil, Roland, SampleCell, audio, WAV, and AIFF. For the publisher, each format has its strengths and weaknesses that extend beyond the capabilities of the machines themselves. For example, publishers must gauge how popular the formats are in the marketplace, where the majority of their sales will be, what kind of library they’re creating, and who will be using it. A number of manufacturers I spoke with agreed that the audio format was the best seller but had the worst profit margin.

“In times past, we did audio discs first and worked toward Mac AIFF,” notes Paul Korntheuer of Rarefaction. “Even though CD-ROMs are superior, people are used to sampling from audio discs.”

Melissa Reuther of Time + Space agrees. “For us, everything comes out in audio first. The CD-ROMs mostly come out in Akai format first, because that format is the leader in the U.K. market. However, the mainstream buyer is buying audio discs.”

Q Up Arts’ Morton says that “certain kinds of sounds are good for certain markets. Generally, we sell big in Japan, and Akai is the king there. We also do CD-ROMs for SampleCell, because we develop in SampleCell. And then, of course, there’s Roland, Kurzweil, and audio.”

“If you’re going to make a business out of it, you have to support all formats,” says Spectrasonics’ Persing. He says that releasing multiple formats simultaneously is one of the company’s recipes for success. “The market is different for each format, so we do different quantities of each. At the moment, Akai is the most popular format on the market. Roland is a good format in the United States because Roland is the only company that has made a serious library of its own. Just because lots of units of a keyboard are sold, it doesn’t mean there will be a CD-ROM market for it.”

It’s no surprise that Persing, as chief sound designer for Roland since 1984, prefers to work in the Roland format. “We will begin a new release in the Roland format, then we’ll do an Akai version. We make our audio CDs from the Roland format, which is the opposite of what other companies do. And because of size limitations, the audio CDs have about half the material that the CD-ROMs have,” he explains.

THE RECORDING SESSION

There are a number of things to consider when making recordings that are destined for use in a sound library. First, you must determine the recording chain, from the microphone to the storage format. These variables are often determined by economics. Perhaps you’ll be able to borrow equipment. If not, what kind of gear can you afford to buy? If purchasing it isn’t possible, how much can you spend on renting it?

Despite the current trend toward higher bit and sampling rates, many of the sound designers I interviewed still record to DAT and work in 16-bit resolution. The common explanation was that samplers and compact disc players are still primarily 16-bit machines.

Of course, you want your master tapes to be in the best-sounding, most robust format possible. If you’re recording in a studio environment, you have higher resolution options, such as the 20-bit ADAT XT20 and LX20 from Alesis, Tascam’s DA-45 HR 24-bit DAT machine, hard-disk recording, or high-quality analog tape. Out in the field, however, the most affordable format is still the portable 16-bit DAT recorder.

For location recording, sound designer Jim Miller uses a portable setup that includes AKG C 414 and Earthworks TC30K mics, and an Oram MWS preamp going directly to a DAT recorder.

Darby’s field rig includes a portable DAT with a Lunatec V2 portable preamp by Grace Design and a Sennheiser MKH 816 shotgun mic. Because much of his work is for film and television post-production, this setup is portable enough to allow him to get into tight places when necessary.

“The recording format depends on the application,” says Ilio’s Mark Hiskey. “Sometimes we go to 20-bit ADAT or 24-bit. Sometimes we go straight to DAT at 44.1 to avoid sample-rate conversion. We’re considering 96 kHz and anticipating DVD as a viable delivery system. The only question is when.”

Daniel Fisher, veteran Kurzweil programmer and director of soundware engineering at Sweetwater Sound, records simple instruments straight to DAT. “For things requiring lots of mics, I’ll use one of the 20-bit ADATs and four different stereo mic combinations.”

Morton does everything on the Mac. “We record directly into Pro Tools/24. Before that we were working in 20-bit for about a year.”

Don’t think that you should put off recording just because you don’t have state-of-the-art gear: ingenuity and recording skill can overcome many deficiencies in your setup. Besides, there are other considerations that will raise the quality and usefulness of your samples.

The majority of the sound designers that I asked record their material flat, preferring to get the best sound possible by using the shortest route between mic and recorder. Mic placement and room acoustics, therefore, play a major role during the initial sessions.

“I prefer a nice, quiet studio with no reflections. Many budget restraints, however, require that you get your recordings where you can,” says Miller. “I wouldn’t add EQ while recording unless there was something extremely wrong with the sound. I prefer to do that later.”

I asked a number of sound designers whether tuning was an issue when recording and whether they used tuners during their sessions. Morton says that Q Up Arts always stresses tuning. “This is the sort of detail that should happen during the original session.”

Others took a more laissez-faire approach to tuning. “It’s never been a problem,” says Miller. “Usually the players will bring a tuner themselves. However, it’s simple to correct tuning in the Kurzweil. You just save the tuning as part of the sample itself.”

“But,” adds Fisher, “the samples must have exactly the same tuning. Otherwise, when you add reverb and switch Velocity levels, you will hear chorusing in the reverb.”

Fisher says that, if there’s a problem, he prefers to “EQ the noise out on the Kurzweil. With the K2500, you can work on a sample and then resample it digitally using V.A.S.T.” He cautions that you have to listen carefully during the recording session. “You have to be insane about noise levels. Pay close attention to ambient room and electrical noises. Remember that when you play multiple notes of these samples, you’re adding the room sound and RFI each time you add a note.”

A final consideration during the recording stage is getting permission to use these performances. Unless you play every instrument yourself, you will need to get written permission from the players to use their sounds in a commercial project (whether on a CD-ROM or in a production). If you intend to license or sell the collection, having copyright-clean samples is a must. Think about this before you contact musicians for the recording session. Be sure to have an attorney-approved contract stating your intentions, and make sure the musicians understand and sign the contract before the session is booked.

FOCUS

It’s easy to spend hours setting up mics and getting the right sound. However, it’s important to conserve your mental and physical stamina for the recording session itself.

For Fisher, preserving attention span is important. “People will spend all day on the miking and sound. But after an hour of recording, you’ll find them reading a magazine rather than paying attention. It’s very easy to have entire ranges of an instrument get wrecked because you lost your concentration for a period of the recording session.”

Attention span is a two-way street. “Be willing to pay your performers,” Fisher adds. “Tell them that this’ll be the hardest and most boring—but most exacting—performance of their life. Let them know that this is a major deal.”

And be thorough in your search for the most useful sounds. The more prepared you are before a session, the smoother it will run and the more you will accomplish.

Before doing a sampling session with the ROVA Saxophone Quartet, sound designer Thomas Dimuzio made a wish list of things that he thought would be useful in a sax-quartet library. “I brought along a list of ambiences, concepts, and feels that I wanted to get on tape. It turns out that ROVA brought an almost identical list. Because of this, we recorded more than enough material for a sound library.”

DETAILS, DETAILS

“Get lots of Velocities,” suggests Fisher, “so that you have lots of choices later. But don’t get too literal about things when programming: if an E sounds better as a D, put it there. Also, remember that the stereo position of two or more samples has to be perfect, because the ear can hear stereo location shifts.”

Fisher is also an advocate of keeping the sampler set up nearby to see how the samples work side by side on the keyboard. “Having the sampler nearby is useful for getting the sounds to blend all the way up the scale.”

“One of the most important things you can do during the original recording is pop the sounds into a keymap and see if they work on the instrument and in scales. Many times, you won’t know that the sounds don’t work until you put them all side by side. Very few people do this. However, this keeps you from getting gaps where the sound isn’t right. Sometimes you’ll find that you need to get a better recording of a single note to even out the scale.

“After you’ve built your keymap, don’t be afraid to throw out a note that doesn’t work and stretch the others to cover it.” Fisher adds.

Miller agrees that having an instrument nearby is a good practice. When he is creating a project destined for the Kurzweil format, he loads the recordings into his K2500 and tries a little programming to get a feel for how the samples will sound in performance. “I like each sample to have a slightly different sound across the keyboard map. When done well, this will give the instrument more of a real playing feel.”

Fisher believes that the more you know about how an instrument sounds, the easier it will be to create a reproduction for the sampler. “When the session’s over, have the musician just play, so you can hear what a real performance sounds like. It’ll give you ideas on how to structure your sample programming. Try to reproduce that performance with the samples.”

Through experience, the professionals also know that how the sounds are used in a library often determines the best way to record and program them.

“Sometimes you need to predict what the sample’s going to sound like in the end; it may not sound the way it sounds in real life.” Fisher points to the acoustic guitar as an example. “Recording the guitar’s open strings and putting open-string samples on each key, rather than getting the individual quirks of each fret, gives the sample a more natural sound. Sometimes a vanilla recording of an instrument is better than one with lots of different clicks and buzzes.”

EDITING AND LOOPING

Now the real work begins. Making many detailed samples from a collection of source recordings is a major task. Whether you are using a looping program, such as Antares’s Infinity, or working in a more generalized digital audio editor, the biggest challenge to the sound designer is seamless editing and looping.

Hiskey says Ilio Entertainment’s first priority is to achieve high-quality sound. “We can’t afford to put out a product that has clicks in the loops.”

Like many Kurzweil programmers, Miller says that he does his work primarily in the K2500 itself. “The way I work now, I go from DAT to Mac, using BIAS Peak to slice it up, and then send it where it’s going to go, which, for many projects, is the K2500.”

Earwax’s Darby tries to keep as many of the samples intact as he can. “Often, they’re from previous jobs, so they’ve been trimmed. Because we use Pro Tools, it’s not a problem to adjust the sound to fit the moment.”

“There are compromises no matter what you do,” explains Dimuzio. “It all depends on the source material.” On Rarefaction’s Etymology, a sample library of guitar and cello sounds played by Fred Frith and Tom Cora, Dimuzio often used crossfade loops. “But some folks may not like crossfades, so I also included chronological loops, where I went in and cut loops from the beginning of the sample to its end. When you place them side by side, you get the entire sample. Otherwise you have its component parts.”

Many elements come into play when you begin creating individual samples. Sound quality, tuning integrity, and level are all things you must consider when choosing the best take. Normalization is an aspect of sample design that Morton thinks important. “Make sure all the levels are hot so that sampler output is optimized. Also, watch that stuff doesn’t get truncated too short. Give the instruments their full decay before chopping them off.”

Fisher suggests that if note attacks don’t match, you should copy and paste the attack transient you like to the one sample that has the weak attack transient. “And make sure your loops don’t change pitch,” he warns. “Your ear is most sensitive to pitch, especially with shorter loops. Moving the end of the loop point will make the difference. Don’t just listen to the sample dry: listen to it in reverb, because, again, you will hear chorusing when both pitches appear together in the reverb.”

Fisher also knows that the ear is good at perceiving when a short loop begins. “If you have a RAM limitation, you’re better off having fewer samples with longer loops than the other way around. Especially with something as complex as a multistring instrument like the piano: you need loops of two to three seconds or more. They should be as long as you can get away with.”

WILL WORK FOR SAMPLES

Many sound-library publishers don’t have in-house recording and editing facilities. That fact means that much of the work on a project is performed by outside contractors working in their own studios (see Fig. 3).

Most of the companies I spoke to expressed interest in hearing about new ideas for sound libraries and collections, especially ones that may already be completely implemented into one of the popular sampler formats. As the market continues to grow, there is also a need for skilled freelance editors and programmers. Because of the amount of time it takes to program a set of samples, you can imagine that many companies have a difficult time keeping up with the demand for new sounds.

“There’s quite a bit of work to be had converting libraries for companies,” explains Spectrasonics’ Persing. “The best part of the project is recording the musicians—making the raw recordings. The next part that’s fun is when it’s done. All the other parts are technical and tedious because of the amount of organization and documentation that’s required. Just coming up with names, it’s hard not to repeat yourself. A lot of thought goes into these details.”

Persing says that to be a good programmer, you need to have full knowledge of what he calls the “arcane archives” of each machine. “If you’re going to make something for the Akai S1000-series filters, you need to know that they respond differently from other Akai versions. Filters are standardized in Roland and Kurzweil machines, but not in the Akai instruments. So, for a certain project it might be necessary to leave the filters alone in that format. You have to be aware of details like how the envelope times work on different models. The more musical you want to make it, the more risky it is. But the rewards are that much greater.”

To Persing, the creativity that goes into the details matters. “Controller mapping makes a big difference,” he says. “If you want to have a filter sweep, a mod wheel or data slider can add in more or less of the effect. That way, folks can tailor a sample in real time.”

Fisher points out the “paradox in creating multi-Velocity-strike samples. You want two different Velocities, but when you switch between samples, the timbre is different. People complain about that. What you have to do is make the crossover points seem invisible. You have to have a separate programming layer for each Velocity range: as you get to the end of one sample range, its brightness should match the darkness of the new layer.”

Fisher also notes that there are reasons to do cross-switching rather than crossfading. “Crossfading rarely works on natural instruments when you’re doing multistrikes. While they’re crossfading, you’re hearing twice as many strings in that crossfade window. With switching, when the Velocity is 0 to 60 you’re hearing the low sample; 61 to 127 you’re hearing the other. Also, when you crossfade, you’re eating twice as much polyphony during the fade.”

But achieving utter realism is not the only measure of success with a sample, says Fisher. “Many instruments that decay do so far faster than your brain thinks they do. So you have to exaggerate the decay time on certain instruments—such as piano, guitar, and bass—so users won’t complain about the sounds being ‘stubby.’ They want a far more linear curve that hyperexagerrates the decay time of those instruments, which is why compression is used so often on those instruments in the recording studio.”

GOING SHOPPING

Shopping your sound library to a manufacturer or distributor is worth serious consideration. Many of the companies I spoke with are interested in hearing new ideas for libraries. “A construction kit really works for us,” says Doug Rogers of East West Communications. “We like to get one- to four-bar loops of all the instruments, then a breakdown of those loops individually, as well as of the individual instrumental notes.”

Korntheuer of Rarefaction says, “I look for collections that are twisted! Something that’s not out there already. Something that spurs the imagination and creativity. If we’re getting a submission, we like to see most of the programming work already done, such as zero-crossovers, normalization, and so on. We also like to have a multilayered sample broken down into its components, so end users can reassemble it themselves” (see Fig. 4).

The process of getting your library heard begins with contacting the company you hope will publish your work. Most of the companies surveyed for this article said that they prefer to get a proposal first before hearing the actual project. If they’re interested in your proposal, they will then request that you send material for them to listen to.

Rather than blanketing companies with CD-Rs of your project, do some thorough market research into which companies are supporting your format and what kinds of products they carry. Look at each company’s Web page to become familiar with its vision. Also, consider what each company doesn’t have in its catalog. If your project is an Eastern European bagpipe collection in the Akai format, target companies that emphasize instrumental collections, world music, and Akai CD-ROMs, but that don’t already have a release of this sort.

JUST DO IT

The market for sound libraries continues to grow each year, as the demand for sound content escalates in the world of audio and video. The result is an expanding market for unique sound libraries and talented sound designers and programmers. With the right sounds and a good presentation, you can carve out your place among the pros.

Gino Robair is an Associate Editor at EM. Thanks to Daniel Fisher, Thomas Dimuzio, Eric Persing, Paul Korntheuer, Jim Miller, Jeff Darby, Melissa Reuther, Doug Rogers, Mark Hiskey, Doug Morton, Jeff Obee, Karen Stackpole, Headless Buddha Labs, Mary Cosola, and Steve Oppenheimer.

LICENSING

When you purchase a sound library, you’re not buying the sounds any more than you’re buying Peter Gabriel’s songs when you buy his record. The sounds on a sample disc are licensed to you for use in your sound productions only. Each manufacturer handles licensing differently, so it’s important that you read the enclosed “licensing agreement” for every collection you purchase.

Typically, the license allows you to use the sounds for your own productions exclusively. Unauthorized use of the collection includes lending, renting, reselling, copying, dubbing, and reconfiguring the samples for resale.

Companies often approach each kind of use of the sounds separately. Record production, film and television, and multimedia projects are each treated in a unique way. Some uses may require a separate agreement with the publisher or may be forbidden entirely. Sometimes you’re asked to credit the sound library, depending on the usage.

Why are sound-library publishers so particular about licensing issues? Consider that each company has spent a lot of money hiring musicians for the recordings and invested countless hours in editing, looping, and programming each sound. The people who do this expect to get paid for their work, just as you and I do. In addition, many companies guarantee that the sounds they’ve licensed to you are copyright clean, so you won’t have legal problems down the line. That aspect of a sound library is an important one that’s often overlooked: for a relatively small fee, these companies are providing you with the tools to continue creating music (and making money) in perpetuity. All they ask in return is that you abide by their licensing agreement.

PIRACY

Every company that I spoke with said that piracy of samples is a major issue worldwide. One company said that the sample-CD industry would be at least ten times bigger than it is now if it weren’t for piracy. That means that the development process of new and unique sample collections has been seriously hampered because of the unauthorized sale and distribution of samples.

It’s certainly fair for musicians to upload their own custom patches or samples to a Web site. It’s another thing to upload something you’ve purchased from a publisher, whether in its original form or modified. We’re talking basic intellectual property rights.

Make sure that you’re not supporting unscrupulous activity when you download sounds from users’ groups or other Internet sources. Any unauthorized reproduction of the sounds, whether posted on a Web site for distribution or copied to DAT for use by persons other than the original user, is harmful to the company that produced the project.

It’s important that you’re aware of piracy issues and respect the work of the manufacturers just as you would like people to respect your own musical creations. If it weren’t for these companies taking financial risks, we wouldn’t have so many fantastic releases in the first place.

WHAT YOU CAN EXPECT IN SALES

Despite the apparent glamour of the sample-CD biz, the companies that publish libraries are not the huge megacorporations that you may think they are. They’re usually small businesses, started by sound designers who thought that others might be interested in their work. While some publishers do much of the work in-house, others commission and license projects from sound designers like you and me.

Before you begin seeing big dollar signs, however, let’s begin with a reality check. Compared with the number of units that traditional music labels sell, sample-CD sales (whether they’re audio discs or CD-ROMs) are extremely modest. Depending on the title, format, style, and price, a sample CD might be considered a success if it sells from 750 to 1,000 units, but the average release sells below that figure.

Selling 600 copies of a CD-ROM priced at $199 may not sound that bad, considering how cheap it is to mass-produce CDs these days. However, when you consider that creating a high-quality library can take up to two years, you begin to see that there are a few challenges in making back your initial investment.

Style-oriented releases (for example, current dance flavor) generally have a shorter marketing life span than their more esoteric counterparts (such as a harpsichord collection). Some sample libraries continue to sell well for manufacturers, sometimes into the thousands, and this kind of success helps the manufacturer invest in new releases to meet market demands.

SAMPLE THE WEB

East West Communications
www.soundsonline.com

Eye & I Productions
www.voicecrystal.com

Ilio Entertainments
www.ilio.com

Q Up Arts
www.quparts.com

Rarefaction
www.rarefaction.com

Spectrasonics
www.ilio.com

Sweetwater Sound
www.sweetwater.com

Time + Space
www.timespaceusa.com