VOICES
by
Todd S. Parkhurst
Delivered to The Chicago Literary Club
October 19, 1998
Voices.
They come from everywhere, continuously, in every pitch and timber, every volume, and form,
constantly, overwhelming us.
Overwhelming us with information information we must have, inquiries we must answer,
demands that must be heard, that must be attended to.
Voices competing with one another, and with other assaults upon your senses for attention.
Voices you don't want to pay attention to. Perhaps my voice tonight.
Perhaps eighty percent of what I say tonight will register with you, and will be understood, and
will be remembered, at least for a little while. But I have no doubt that some amount of what you
remember of what I have said will fade, starting soon, as I continue to read my paper. You will
begin to think of other things family, work, fun, the trip home, the just-finished dinner, the
person next to you, some other interest. So, soon, perhaps fifty percent of what I say will register
with you perhaps less. If you become angry with me, the percent of information registering will
drop to ten percent, as you think of retorts or rebuttals. If you fall asleep it's been known to
happened at the Club the percentage ratio of information broadcast to information received,
processed and stored or remembered will drop to zero.
What must I know, what must I do, what techniques can I use, to communicate most effectively
to you? To grasp your attention? To hold your attention? To get you to listen to me? To
understand me? And, I hope, to agree with me? Given the Literary Club's formal restrictions on
my presentation I cannot show a movie, or have actors engage in dramaturgy -- what can I
do?
How can I use the only instrument at my disposal my voice? How can I maximize the
effectiveness of my voice communications?
I shall discuss tonight two of the most difficult environments of human voice communication I
know. The first, mouth-to-ear communication is as old as the human voice itself, but modified,
disembodied over the past seventy-five years; revised, restructured, reconsidered again and again.
It is an environment requiring expensive but mostly invisible equipment. Here art rules, and
technology simply conveys the art. And you, my audience, quite consciously do not want to hear
the voices, and do not wish to attend to the art.
The other environment is really no more than one year old, practically speaking, and is as
technologically intensive as you can encounter. You stare at the equipment. The user uses the
equipment in a personal way, but the equipment is relatively inexpensive. The technology requires
substantial further improvement, which surely will occur over the next few years. It is beginning
to, and will, completely revolutionize writing. It's mouth-to-eye communication. Strict speech
techniques are essential to efficient operation. Every word ought to be enunciated the same way
every time. Art, here, is almost entirely subservient to technology.
The first environment, or art form, is one you encounter every day, like it or not. Mostly, I say,
you do not. You tune it out. It is the art of radio and television commercials. The performer
the talent, he or she is called is the narrator, the announcer, the actor, the voiceover.
The voiceover person may have the most difficult job in all the modern world of communications.
He (or, increasingly, she) must grab your attention; hold it; interest you in a product or service or
idea, induce a positive if subconscious impression in your mind, and leave you with a clear mental
image which you will remember all in thirty seconds. Or less.
Here is an example of the problem the television and radio advertising community face: The
Dubuque Supreme Meat Company of Dubuque, Iowa, wants to increase the sales of its meats. It
wants its products more widely known. It wants to make people want its products. It is willing
to spend considerable sums on advertising but it wants results. It wants measurably increased
sales for the money it spends. What is the quickest way to engage the widest audience, or set of
potential customers, with the ideas that these products exist, and are attractive? Perhaps radio.
Perhaps a 30-second spot would work:
Life's weird. Iceland is green. Greenland is ice. And how much longer are they
gonna call New York new?
At least Dubuque Supreme meats really are supreme. They're tender, juicy, and for
a limited time, you can try one free. Just buy a package of Dubuque Supreme Pan Size
Bacon, and we'll send you your money back.
So, the Red Sea is blue. Dubuque Supreme meats really live up to their
name.
How does this ad get made? How is it directed and produced so that it is most likely to be
effective that is, most likely to attract customers and build sales for Dubuque? That's art.
That's voiceover performance art. And Chicago is one of the largest and most important markets
in the USA for this voiceover work. By some measures, the Chicago market is larger even then
the New York market. Many highly trained voiceover artists live and work in the Chicagoland
area
In general, voiceover work is the reading, or narration, of TV and radio commercials; industrial
films or narrations; and, increasingly, instructional or same CD-ROMs. Some 65,000 voiceover
jobs are offered and accepted each year. Some very few and fortunate individuals obtain so many
jobs and are paid so well for each of them that their incomes are estimated to reach high into the
six-figure range. Many others are able to earn a much more modest, yet comfortable, living.
However, the vast majority of voiceover talents earn anywhere from a few hundred dollars per
month to perhaps a few thousand dollars per month doing voiceover work. It's great work for
people working for themselves at home; for at-home moms; for recently retired people; and for
others who can run over to a recording studio on short notice. Obviously, the more time you
spend at promoting yourself to agents and the more casting calls you can make, the more jobs you
may get. Not surprisingly, you ought to work at this full time if you expect to make full-time-career money.
First, consider the writing, or copy, for this advertisement. It certainly is not Shakespeare, or
even Tom Clancy, but undeniably it has been carefully prepared. In just 30 seconds, the radio
voice commences some mild but arresting, slightly quirky humor, introduces the listener to
Dubuque Supreme meats, describes the products in an attractive way, suggests a reason to
purchase them, and returns to the humorous theme to wind up the ad. The ad copywriter needs a
keen ear for humor, and for words which will create a relatively powerful, memorable mental
image in a positive way.
Notice, please, the lessons here for Literary Club papers: to arrest and retain attention, use short,
simple sentences. Use memorable phrases. Make image-producing word choices. Suggest the
order of importance of ideas by word sequence, not subordinate clauses. A careful word count
and syllable count produces an oration of exactly the desired duration.
Having decided upon the general theme for the advertising campaign, and having developed the
particular radio spot advertisement script, the advertising agency contacts various voiceover talent
agents, and asks those agents to suggest individuals who might be attractive candidates to record
the script. Each agent contacts those individuals whom the agent thinks might be best- suited and
most likely to be cast and asks each of them to audition for the job.
The audition may be held in a recording studio in Chicago, or, increasingly, it may be conducted
over a virtually noise-free telephone system. These new phone systems consist of a microphone
and studio, perhaps located in the individual voiceover artist's home, which is connected to the
main studio or other receiving facility by a telephone line called an ISDN line.
Larger clients and advertising agencies usually want to audition and hire union talent; that is, they
ask that the voiceover artist be a member of the American Federation of Television and Radio
Actors or the Screen Actors Guild. Union work pays something on the order of three hundred
dollars for the first hour of narration work, and perhaps one hundred dollars for each half hour
thereafter. Non-union work can pay two hundred dollars to, say, five hundred dollars for an
entire narration job. There is more money in spot advertising work. The agent's fee may amount
to 10 percent paid to the agent by the voiceover artist, and another 10 percent paid to the agent
by the advertising agency. Smaller ad agencies and smaller clients are more likely to work with
non-union talents; the pay scales are less.
Now, as you might suppose, people do not just wake up one day and decide to become voiceover
artists. Months, even years, of training have become nearly essential to break into and succeed in
the voiceover business.
In virtually every case, the talent agent has received and carefully listened to a tape, called a
demo, of the voiceover artist's work. The preparation for making a demo tape can take six
months or more of weekly or sometimes bi-weekly sessions between the aspiring voiceover talent
and the voiceover coach or producer. Ideally, the demo tape is a 2-1/2 minute cassette of the
voiceover artist performing short samples of scripts which display the artist's range of styles. The
segments may vary in length from four second tag lines (for example, the slogan at the end of a
spot, like Microsoft's Where do You Want to go Today? Or the old when Better Cars are Built,
Buick will Build Them). A segment might run up to 18 seconds for dialogue. Perhaps 8 to 12
samples will appear on one tape. The tape must demonstrate the range of styles which the artist is
able to perform excellently, and for which the artist could reasonably expect to get cast. For
example, the demo tape may demonstrate a so-called in-your-face attitude. Or the tape may
demonstrate a warm, confidential, atmosphere and personality. Perhaps the tape will demonstrate
an authoritative, reliable voice and persona.
During the coaching in preparation for the making or producing of the demo tape, the best
coaches constantly evaluate the student talent. If it appears to the coach that the student is not
making sufficient progress, or if, for any other reason and at any time, the voiceover coach comes
to believe that the student talent does not have a realistic chance of succeeding in the voiceover
business, he/she will politely decline to continue the coaching.
This demo tape constitutes an investment in the student voiceover's future. Accordingly, it must
be of the highest quality, both technically and artistically. Because this business is so highly
competitive, only tapes of the very finest artistic and technical quality will be seriously considered
by voiceover talent agents. The tapes must include interesting scripts and state-of-the- art
production values. A great number of the copies of tapes must be made, labeled, and packaged in
cassette boxes. Interesting, memorable, packaging cards, called J-cards are essential. Ideally, the
tape J-card will provide some memorable visual guide to the voiceover artist. One I have seen has
a cartoon of a cerebral-looking guy in a lab coat. The caption says, "You don't have to have a
rocket scientist for your voiceovers. But why take chances?" Pat Byrnes was once a missile
propulsion engineer.
Ideally the scripts are scripts of actual commercials. Ideally, they will be written in a creative,
interesting and attention-catching way. The business is dominated by twenty-something
generation x-ers, and so the jokes, catch-phrases and interests of that age group dominate the
commercials, especially for products offered to young adults.
Good voiceover talent, and good talent coaches, have an almost abnormal sensitivity to human
voices. These people instinctively sense subtle shades of meaning and feeling never recognized by
the more normal, more pedestrian, mind. Indeed, some voiceover coaches approach coaching as
a mystical, or religious, calling. These coaches are able to sense, or pick up, creative blocks
which the individual voiceover artist, or student artist, may be personally unaware of. Sometimes
these creative blocks constitute worries with which the voiceover talent must contend in his or her
own life. For example, it would be extremely difficult to do your best voiceover work if your
child was seriously ill. The speaker who is physically tense may be inhibited, even if ever so
slightly, from opening the mouth properly, and so part of the voice may go up into the nose
instead of out through the mouth. Perhaps the shoulders will be slightly hunched and so the chest
will not resonate in the usual way and the voice gets caught in the throat. Relaxation techniques
such as yoga may loosen muscles and consequently alter speech, improving the sound of the
voice. Coaches also work with the students on particular ways of phrasing which may be
individual to the voiceover student and may become counterproductive if over-used. Normally,
the goal for the voiceover artist is to be anonymous. Usually the advertiser wants to sell his
product, not the actor or his voice, but occasionally the client wants a recognizable voice to lend
authority and credibility to his product or service. The alarming, dangerous sounding voice of
the famous movie actor Jack Palance can now be heard on local radio urging everyone to come
see the musical Ragtime.
Good readings by good talent require good understanding of the script, the client, and the
product. The voiceover talent must know why the dialogue in the script is true. The voiceover
talent must know why the messages are important. Voiceover talent must know how the product
or service works. The voiceover talent must know the problem which the product or service
addresses, and how what is promised in the script will solve or better the problem.
The voiceover talent must know who he is in the script: is he a person with a particular job? A
particular attitude? A particular age? Is he an announcer? Is he an important character?
Moreover, the voiceover talent must know something about the opinions and feelings of the
character he is playing in this 30-second speech. And the voiceover talent must also know who he
is talking to: who is his audience? How can they best be engaged? How can they best be
persuaded? The voiceover talent must know why he is saying these things in the script to this
person in the audience. When it all goes together right, it can sound like this:
At Walmart, we always greet you at the door...
we always try to have what you want...
you can always return it...
and you will always find the low price on the brands you trust.
Good voiceover artists --whether they are beginning students or experienced practitioners --carefully read over the script several times before attempting to deliver it. These individuals then
read aloud the script in the assumed performance voice two or three times. Mastery of oddly-worded phrases or tongue twisters is accomplished here. Run-on sentences are identified and
marked with a pencil. Next, the artist will mark the script, using his own shorthand, with
reminders as to where to breathe, what words to stress and how to stress them, what phrasing to
give a group of words, and relative highlighting or importance between phrases and individual
words.
Clues to this phrasing and stress are found in the way the script is written. The may also be
found in the client itself: in the accompanying music or visual scenes; or in other materials.
During this practice, and certainly during the actual recording performance, facial expressions are
used, hand gestures are used, and posture is varied to help provide just the right tone, inflection
and pacing to the voice.
Tone, shading and attitude can also be varied by the reader's physical and mental relationship
with the microphone. "Talking past" the microphone in this way can provide a breathy, intimate
atmosphere or feel. Addressing the microphone directly provides a more normal, conversational
atmosphere. Sometimes a loud, raised voice is required, but yelling at the microphone may
damage it. Accordingly, the voiceover artist may need to distance himself from the equipment
like this. Turning the mouth just slightly away from the microphone lessens or eliminates
annoying plosives, such as those which can arise in "please pass the peas". It minimizes the
sibilant esses, as in "sassafras suits silly Sally".
Accents can provide color or atmosphere. The increasing diversity and sophistication of today's
audiences mean that imitation accents can be used only for comic effect, and then only with
extreme caution. An obviously-artificial Hispanic or African-American accent may well outrage
listeners. A talent to whom such a voice or accent is natural will, these days, inevitably receive
the nod for the job. There is a certain school of thought that the middle-American accent heard
in some portions of central Ohio and in the Northwestern suburbs of Chicago constitutes the
quintessential American accent. Voiceover artists who, knowingly or unknowingly, offer this
accent and speech patterns are in demand.
Women's voices have long been used in voiceover work, but women now compete for almost
every job even to advertise products which are of almost exclusive interest to men. Picture this
TV commercial: a very trim man and a woman are standing close to one another outdoors,
exchanging loving glances.The woman's voicevover says:
You know, I love my husband. He's bright, witty, makes good money. And he has
the body of man 10 years his junior. In fact, he's everything a woman could want.
Six months ago he bought a new Skeeter fishing boat. And if he'd ever drag his
buns out of it and come home, I might introduce him to my new boyfriend Bob
here.
A considerable proportion of the radio commercials offered today make use of humor, or quirky,
attention grabbing ideas. An effective format for this is dialog. Dialog of this sort requires two
or more very different voices having highly contrasting pitch and inflection, but a rapid-delivery
pace is required. Voiceover actor 1 must begin to deliver his line just as voiceover actor 2
finishes her last word or syllable. The rapid-fire delivery arrests the attention of the listener.
Moreover, the voiceover talent audition for script must know against whom he is competing.
For example, a 57-year-old male competes against many highly-talented, highly- experienced,
voiceover artists who have been in this business for 10,20, 30, or even more years. Edward
Herman, who played Franklin Delano Roosevelt in a PBS biography, now stars in Dodge truck
and car commercials. Don Pardo of the soaring phrasing, has become so popular once again that
a small legion of copyists has developed. The advertising agencies and talent agents know these
individuals. The advertising agencies and the talent agents are much more likely to hire a
"known quantity" than a relatively unknown.
Once hired, the voiceover actor can expect to spend one or two hours or more in a studio,
recording take after take of the 30 second spot script. The commercial director will ask for
quicker pacing, now slower, now more emphasis here, now lessened emphasis there, a more
drawling, country voice, now a more urbane, flatter sound. After most of the possibilities and,.
It sometimes it seems all of the personnel--have been exhausted, the sound engineer goes to
work. Some of these sound engineers are truly creativity geniuses. A good sound engineer will
seamlessly splice a sentence from take one with a phrase from take five, cut a syllable from but
then use a word from take seven, and then add background music sometimes especially
composed and played for that spot to produce the final cut, or work, on a digitally recorded
tape.
That tape is then reproduced and distributed to radio stations who are paid to broadcast it.
Because voiceover work is performance art, variations are encouraged, and no two takes are
ever the same. But in another voice environment, perfectly identical renderings are best.
Perfectly identical renderings are most efficient. Perfectly identical renderings produce the most
effective results. This is the environment of speech recognition software for personal
computers
Continuous-speech recognition software is a relatively young technology. A small software
company named Dragon Systems introduced the first general-purpose continuous-speech
recognition program for personal computers in June of last year. It was called
NaturallySpeaking software. IBM Corp. followed soon after with its ViaVoice software. That
software is now being extensively marketed in a television advertising campaign which uses on-
screen actors, and voiceover talent. You have, I'm sure, encountered the tag line "You talk. It
types." More recently, Lernout & Haspie has introduced its Voice Xpress software, and others
have entered the field with competing products. All these products have been intensively
upgraded and improved during just the past few months.
As you may suppose, the accuracy of the speech recognition program is essential to the effective
and efficient use of this technology. If a program does not accurately set forth the words you
expect to see when you dictate, the program has no advantage over the usual and classical form
of word input typing. Program effectiveness evaluators suggest that NaturallySpeaking is, by a
small margin, the most accurate speech recognition program currently available, followed closely
by the IBM product, and by Voice Xpress. After some initial training --the user trains the
software, not the other way around -- NaturallySpeaking software will accurately recognize 95
percent or more of the words spoken into a microphone connected to a computer.
Most of these programs can be easily coupled to any of the major word processing programs.
For example, I use the NaturallySpeaking program with the WordPerfect word processing
system in my computer. To prepare this paper, I simply put on a lightweight headset having a
small microphone, started the NaturallySpeaking program, and began dictating into the
microphone. After a very slight pause, the words appear on the computer screen formatted in
the WordPerfect program. I have found that a good strategy is to dictate a few sentences, or
perhaps a paragraph, and then check the just-created text for errors. For example, only three
small errors, all easily corrected and identified, occurred during the dictation of this paragraph.
It is easy for old-time lawyers, such as myself, who have been using dictating equipment for
many years to put out good first drafts of documents with little or no hassle.
When you begin using direct dictation software, you will soon discover that you can prepare
your written work product much more rapidly and more accurately using this new technology,
unless you are an extremely skilled typist. These new direct dictation systems never, ever, make
spelling errors. Their sense of context is still lame, and they occasionally make hilariously wrong
word choices, but they put words on the screen and into the word processor far more rapidly
then most non-secretary amateurs can type. (For example, this system momentarily believed I
was discussing "world chess," not word choices.)
And the system becomes more accurate with increased use. It "learns" the particular inflections,
pacing, tone, and speaking style of the user. The system notes the corrected correlation between
your voice (really, the electrical signal generated by the microphone ) and the word chosen.
Over time, the old common mistake becomes rare. Each of these programs have a built-in
original vocabulary of between 50,000 and 64,000 words, and new words can be easily added
and "taught" to the system. For example, my system has learned the words "biocompatible," and
"Broeksmit," even though those words were not in the original system vocabulary.
"Biocompatible" is still not found in the WordPerfect word processing system spell checker
dictionary, and even I cannot spell "Broeksmit" without looking it up.
So these direct-dictation software systems are accurate, and they become more accurate with
extended use. While they will never become perfect, the nigh-on universal consensus is that they
will become so accurate within the next few years that secretarial copy-typing of extended
dictation to create long articles, memoranda, briefs, or the like will become virtually extinct.
Certainly computer keyboards will still be necessary. They provide the ultimate ability to make
corrections. They permit difficult-to-dictate words to be inserted, and complicated changes to be
made. Secretaries will not lose jobs. Indeed, secretaries will become even more important as
editors and custom publishers of written word product.
The developments in accuracy and learning in these programs will tend to make them
increasingly attractive to the general public. Like much of the general public, I'm not a computer
whiz. I quite agree with Walter Mossberg's column in the Wall Street Journal that "the personal
computer remains the only common possession that makes smart people feel stupid and requires
the constant ministrations of a priesthood of experts." Unlike the telephone, television or fax
machine you use, your personal computer requires constant upgrades, and behaves erratically,
introducing a new hassle or two for every one it supposedly eliminates." But I really do believe
that Microsoft and other software developers are trying to make things easier for us poor end
users, and I believe they will succeed.
One of the ways computer program designers will succeed is in the area of integration, and that
is happening in voice dictation software, too. I can issue operational commands to my
computer, as well as dictate text. I believe the time is not far off-no more than a few
months when I will be able to simply say "computer wake up. Open ACT! Look up last name
Broeksmit. Write letter, personal letterhead. New Paragraph. Thanks for having me deliver my
paper entitled "Voices" to The Chicago Literary Club. As you requested, here is a copy of my
paper for the Newberry Library archives. Standard close. Print. Print envelope." Immediately
the computer printer will deliver the letter, in condition for my signature complete with a
properly addressed envelope.
Or I will be able to dictate that letter into an E-mail message that can be sent immediately to
Jack's computer system, saving the paper and time of the old snail mail.
As I indicated earlier, enunciating the same words in exactly the same way every time provides
maximum accuracy and system efficiency. But still and all, it's never going to be as much fun as
the famous delayed sign--off of Paul Harvey: "... Good Day!"