Posts Tagged ‘IVR’

Why Can’t I Sound Like a Human?

Any voice talent reading this post will probably agree with me: there is no other more commonly heard direction — either from engineers directing a voice-over session or the clients and ad agency writers looking up from the cheese platter and offering their suggestions on how the spot should be read — than the following:

“Just sound like a real person!”

Doesn’t sound too labor-intensive, does it? After all, I *happen to be* a real person, who on a daily basis tells the barrista how I want my coffee; who pleads with the grocery clerk not to pack all the heavy groceries in one bin (they don’t listen) — and yes — I have real, natural conversations with friends and loved ones on a regular basis.

When we step in front of the mic — whether it be for a broadcast spot, an industrial film voice-over, and yes, even IVR prompts — something clicks in our brains and we default into the thinking: “I’m working. I’m a professional voice. Therefore, I must speak professionally.” The ad execs can’t really mean it when they say they want you to sound like their receptionist — otherwise they could have just dragged her here in the front of the mic — and for a lot less money, right?

The trend — especially if you listen to TV voice-overs — is candid, natural, “everyman”.  Almost gone are the days of a slick, bass-y male voice tantalizing you with talk of V-8 engines and Anti-Lock Brake Systems — many car ads now feature voices which sound so completely accessible, and for lack of a better word…..ordinary, that you don’t feel like you’re actively being sold a car (it’s a trick), but rather, the announcer just sounds like your neighbor, responding to your question shouted from the next driveway over: “So, how do you like your new Mazda?” The voice for Wendy’s sounds not unlike the voice you’d hear thanking you, as she hands you your burger at the take-out window. About the same age-range, and roughly the same amount of “polish”.

We almost have to consciously let go of some of our experience and training, and approach the material as through we’re seeing it for the first time, *saying* it for the first time — and — this is key — that we don’t have the nicely modulated voices or clear diction on which we built our careers.

And that’s not a problem — if the *material* itself is *written* in a conversational tone. There’s nothing more pleasant than being cast in a quaint two-hander radio spot that’s written with how real people talk in mind — and still manage to sell the product. The writer has been mindful to write sentences which might realistically be said between two humans. However, all too often we run into danger areas when “Marge” says to “Celeste”, for example: “Well, Celeste, Effexor isn’t for everyone. Oh, no. People who are prone to Tachycardia, Hepatitis, Chrone’s Disease, COPD, Osteoporosis, or if you have any of the following; changes in mucous color, increased cough, blurred vision. Certain people shouldn’t take Effexor: people with high blood pressure or lower bone mineral density. Do not take Effexor if you suspect you may be pregnant.” (They coyly cover their mouths and giggle.) If the material is at least written with a conversational “ear”, we, as voice talent, might have a reasonable chance in translating that into candid, natural conversation. I was assigned an on-hold script about a year ago — with the direction: “Sound Like a Real Person!” — and the material dealt with marine-grade sealants, industrial lubricants, and all manner of sewage interceptor and collection lines. Horrifically dry and technical content. “Just imagine you’re saying this to your best friend!” came the direction over the phone patch. You know, whenever I gather the girls together, and the good martini glasses come out, talk will invariably turn to the debate between LPS1 Industrial Lubricant and it’s rival, Mobil SHC. We’re still fairly divided about that issue. Don’t get us started.

In the voicing of IVR prompts, the challenge to sound natural becomes even more important — and arduous. Given the automated nature of telephony prompts, the “sameness” required in order to make the prompts flow effortlessly together requires a steadiness in inflection — and doesn’t exactly invite creativity in the voicing of the prompts. Even if you have a wonderful, relaxed, conversational opening prompt: “I can help you find what you’re looking for. Why not tell me more about what you need? ” you are still at the mercy of robotic-sounding numbers, months, and other “set” landmarks built into your IVR system. I try my best to sound as “real” as possible — and material that is written in a relaxed, conversational tone helps your voice talent to also express that naturalness audibly.

Almost like a model who is hired for a print ad in which great time and resources are used to “uglify” her with mud and dirt, so should voice talent realise that there are times when you’re hired for your melodic tones and crystal-clear enunciation — and other times where all the polish and refinement needs to be stripped down and to access that “everyperson” voice. You know the one. You use it each and every day.

Next blog: I”ll write about the challenges of accommodating clients who require accents…..great and fun work, if it’s done well!

Thanks for reading. Feel free to comment!

It’s Broke — And We Ain’t Fixin’ It

The link to a New York Times article was sent to me a couple of weeks ago by a client of mine — uReach Technologies, out of Holmdel, NJ ( — they supply voice mail service to many telephony companies, among many others, Verizon. They’ve hired me for years to voice not only platforms for Verizon, but for numerous other carriers. Here’s the link to the NYT article:

The author, David Pogue, is a well-known technology writer, an emmy-winning technology correspondent for CBS News, and the leader of the “Take Back The Beep” campaign, inspiring some of the major carriers to — in some cases — completely eliminate the 15-seconds of instructions before leaving a message, or at least greatly reducing the instruction wait time with other carriers. 
In his article, he mentioned being sent an e-mail from a marketing executive at uReach; the title of the press release being: “Ninety-Eight Percent of Users Want a Better Way to Get Voice Mail” — echoing Pogue’s own frustration with the mind-numbing (and revenue-generating) wait, listening to the same old instructions. 

Why should I — a person whose bread and butter is voicing voicemail trees (and the longer and lengthier the better!) — applaud Mr. Pogue’s efforts to reduce the laborious and time-consuming IVR trees? Those who have been following this blog since its inception last Fall know that I am an evangelist of “Keep It Simple.” Get to the point; say what you need to say in as few steps as possible; don’t underestimate how short people’s attention spans are (mine included), and even though a lengthy voicemail tree generates money (Pogue estimates: “If Verizon’s 87 million customers leave or check messages twice each business day, that comes out to $750 million of airtime a year..”) realise that listening to long menus is a collosal waste of time and productivity.

Mike Comstock — my contact at uReach and the source who brought the atricle to my attention — sent it to me in a sprit of “See why your scripts are so wordy?”, but Mike had some great further thoughts on the issue, especially the aspect of supplying an awesome technology, which gets hidden behind an old interface:

“I like to refer to the interface as a skin,” Mike explains. “If a web page looks unattractive, cluttered, confusing, etc, then it doesn’t matter how cool the underneath technology is. If the skin is ugly, then customers aren’t going to stick around to discover the cool stuff underneath. Similar parallels hold for the IVR skin. No matter how professional you or any other voice talent may be, if the script/call flow is poor, then the product suffers.”

There’s other voice talent? I digress.

We’re all after the perfect blending of an outstanding technology that drives a system which is a dream to navigate aaround; never frustrating, not delliberately trying to bill more minutes by asking callers to listen to Herculan menus; hopefully, a soothing, professional, genuine-sounding voice provides the icing on the cake and creates a lasting impression and even a company identity.

Join me here for my next blog, where I will investigate a conundrum which had plagued professional voice talent since the birth of sound recording: Why is it So Darn Hard To Sound Like a Real Person?


I’m (Apparently) Too Sexy for This IVR

A few years ago, I voiced an interesting project for Kodak: a self-serve photo development kiosk where photographers could process their own film by simply walking up to a kiosk, installing their film, and controlling the processing of their photos in a very turnkey way. I was actually hired by a sound studio in Australia do voice the “North American-Sounding Female” version — the idea (especially in retrospect) became antiquated pretty fast with the advent of digital picture-taking making direct-to-film photography all but obsolete to all but a few hobbyists. Nevertheless, the project went forward, and with the studio being happy with the sound files and the invoice paid, I didn’t think too much about the project — the product hit the stores, and it promised to be a huge time-saving (and more private) way of having photos developed — however short-lived.

 A couple of months after the kiosks were up and running, the sound studio sent a very carefully-worded e-mail to me: in the sound files I submitted, I was sounding too…sultry. Breathy, languid, kitten-with-a-whip. Further to that point, marketing researchers from Kodak were tracking the film processing stats closely; male customers were coming in droves with their rolls of film, agreeably following all of my prompts, and leaving generally contented. Women, on the other hand, were beginning the process, but soon cancelling out of the system and leaving the stores with their arms crossed and scowls on their faces. The feedback that Kodak got: I was sounding just a little too….suggestive.

I had to be honest with myself, and acknowledge that I had actually gotten that feedback from *other* clients on occasion, ever since I started voicing professionally. It’s not anything I meant to send out; it wasn’t a deliberate “device” on my part — it was a certain bubbly “friendliness” that was coming across unintentionally as, well….come hither. Regardless as to whether or not it was deliberate, it soon to became clear that the “breathy”, somewhat heady treatment of sound files I was doing wasn’t going to fly — unless specifically called for. Redos were done in a more businesslike-yet-friendly tone, which proved to be much more accessible to consumers of both genders.

One must realise that  — even honoring the time-tested truism of advertising that “sex sells”, we must always keep our audience in mind, and that at least half of that income-earning and decision-making audience is female. I needed acknowledge that the tone and approach of sound files — be it for telephone systems, streaming audio on websites, or even advertising — must appeal to both genders, and that an overly-sensual female voice will invariably alienate women — and no doubt the converse would be true, with men tuning out a male voice who is trying too hard to suavely win over the women.

I’m definitely asked to put on a provocative tone for the occasional customer — the IVR for the makers of the Ferrari Art Engine comes to mind, where “the more languid the better” was the direction I got; there are other clients — such as the one who hired me to voice the IVR for the US Prison System — who urged me to avoid at all costs anything even close to an inviting tone. Otherwise, a friendly, professional, non-threatening, accessible tone almost always seems to work for most projects — and keeps me out of trouble.

Next blog: I’ll discuss my favorite types of clients…those clients which do everything right and make my job a pleasure!

The Basics of VoIP

Don't Worry -- Your Fingers Still Do The Walking

When I first started voicing the prompts for Asterisk, I must confess to having absolutely no knowledge about VoIP whatsoever. I only knew that VoIP (Voice over Internet Protocol) — of which Asterisk was a shining example — is a technology which enables voice communications over IP networks, such as the internet or other packet-switched networks, rather than the traditional Public Switched Telphone Network (PSTN), typically used by the old, pioneer telcos.

That explained all the T-Shirts I kept seeing at the first Astricon, with the Bell trademark inside a red circle with a line through it.  (I thought it was an anti-phonebooth movement…)

Boiled down, it works like this: an Internet telephone call is a conversion of the analog voice signal to digital format and a compression/translation of the signal into Internet Protocol (IP) packets for transmission over the internet. The process is reversed over the reciving end. Easy Peasy!

While the first glimmers of the germination of VoIP can be traced back to 1974 a paper published by the Institute of Electrical and Electronic Engineers entitled “A Protcol for Packet Network Interconnection”, (it’s on my night stand — really) it wasn’t until roughly twenty years later that  mass-market VoIP services over broadband Internet became available, enabling inbound and outbound calling, unlimited domestic calling, and to some other countries as well, for a flat monthly fee and free access to other subscribers using the same provider. Amazing savings and freedom for the calling public — with unheard of bandwidth efficiency and the low costs that VoIP can provide, businesses soon embraced the VoIP movement readily. Throw into the mix that VoIP has evolved into “Unified Communications” — treating all communications: phone calls, voicemail, e-mail, fax, and web conferencing; as discrete units which can delivered to any handset — including cellphones….it’s an evolution in communications which has been exciting to watch, and will continue to amaze.

There have been glitches. Quality of service — particularly, delays and latency (caused by the physical distance that the packets travel), can be greatly alleviated by relieving the congestion by means of teletraffic engineering. The issue of emergency calls dropping out — due to IP making it difficult to locate networks users geographically — has been a massive challenge technically, and even legally. Factor in the mobility that IP allows (which, remember, is a benefit) — the IP address has no relationship with a physical location. Big problem if you’re hoping the ambulance will find you and remove the whale harpoon lodged in your thorax. The VoIP E911 system — which associates a physical address to the calling party’s telephone number — fulfills the requirement of the Wireless Communications and Public Safety Act of 1999. However, it’s only as effective as the information it has: subscribers must be dilligent in keeping their address information up-to-date.

Security is another area in VoIP has some fragility: VoIP systems are susceptible to attacks, as are any interconnecfted devices. Hackers can create denial-of-service attacks, harvest customer data, break into voice mailboxes, and even record conversations. Maintaining security and still allowing VoIP to traverse firewalls continues to be a challenge.

Ease of communication. Affordability. Accessibility. Reliability, Not just buzzwords in the telecommunications industry — they’re the cornerstones of VoIP, which is indisputably the present gold standard of communication, and most definitely the future.

Next week’s blog: tag along with me at the IT (Internet Telephony) Expo convention in Miami, and watch the schmooze habits of the Voicegal! Many demos will be handed out! There might even be some on-the-spot outgoing messages recorded on prospective client’s cell phones….all in the name of marketing, and convincing them that mine is the voice they need on their systems!

The Telephone Lady Speaks

My name is Allison Smith, and I have — to say the least — an unusual job.

Most people listen to automated voices on telephone systems on a daily basis and never really give much thought as to just *who* does those recordings — or even that they are articulated by an actual living being.

Well — I *am* that being — or one of them. I have been described as one of the most prevalent telephone voices in North America, having voiced platforms for Verizon, Qwest, Cingular, Vonage, Bell Canada, Twitterfone, and Asterisk — an Open-Source  VoIP platform that has the honor of being the fastest-growing telephony platform in the world.

Headshot With Phones

It was brought to my attention by a valued client that at any given time, my voice is playing in someone’s ear — somewhere around the world.

I work for a vast clientele globally, in a wide assortment of industries, and it seems that the applications to which my voice can be applied is limitless. Visit my website at for a full sense of my client base — and next blog, I’ll write about some of the more…exotic, strange, and just plain wierdball requests I get.

Thanks for reading!