There are over 700 million iPhones in the world, and since 2011 they’ve all come with Siri, a virtual personal assistant who can help you do everything from check weather forecasts to drunk-dial your ex (hey, she’s here to help, not judge). In America, Siri’s voice was provided by Susan Bennett, in a role that catapulted her from successful but obscure voice actor to slightly more successful and slightly less obscure voice actor. (That’s about as much as voice actors can strive for.) Susan told us all about the weird lessons she’s learned from having her voice come out of everyone’s pocket.
Apple puts more thought into their product packaging than some companies put into their products, so we assumed the auditions for Siri would be a rigorous and highly competitive process. But while we’re sure Apple put a lot of careful thought into Siri’s code, Susan had no idea they even used her voice until early adopters started asking her which Carl’s Jr. was least likely to give them food poisoning:
“I didn’t know it was the Siri gig; I didn’t realize anyone was auditioning me for it. I found out I was Siri when Siri launched. A fellow voice actor emailed me and said, ‘We’re playing around with this new iPhone; isn’t this you?’ And I was just like, ‘What?'”
No, Apple didn’t secretly record Susan in her sleep (at least as far as we know; maybe they have since). Siri was developed based on recording sessions Susan did with another company. Susan had done a lot of interactive voice response work (the voice on the phone that tells you to press 1 for English) and thought she was just doing more. “The recordings were done for a text-to-speech company starting in 2005. Apple got all their Siri voices from this company, but they came in after the fact. We had no contracts with Apple.”
It’s bizarre that Apple didn’t feel the need to even give Susan a cursory, “Hey, just FYI, we bought your voice, and soon millions of teenagers are going to be trying to trick you into saying dirty words.” So Susan’s feelings about instantly becoming one of the country’s most recognizable voices were … mixed:
“I had really ambivalent feelings. I was flattered to be chosen to basically be the voice of Apple in North America, but having been chosen without my knowledge was strange. Especially since my voice was on millions and millions of devices.”
But the biggest issue was money. Susan had no contract with Apple, and that means she hasn’t seen a dime beyond what she was paid for her initial IVR gig, despite being a key part of one of Apple’s big selling points:
“There were Siris all over the world because, for instance, I don’t speak Thai or Japanese. They had to have native speakers. And all of the original Siris weren’t paid for the usage. We were paid for the original recording sessions, but we weren’t paid for being on all those phones. Which is a pretty big issue for us. I know there was one person who ended up getting fired from another job because he was working for a company that considered Apple a competitor. His voice was on the iPhone, and he lost a job because of it.”
Susan’s not saying she should be driving a Lamborghini painted iPhone-white, but considering Apple has enough money to overthrow some countries, they probably could’ve afforded to cut her a decent paycheck. So, the next time Siri reminds you about your proctology appointment, keep in mind that she’s doing that dirty work pro-bono.
The Recording Sessions Are Long Hours Of Painful Nonsense
The logistics of most voice acting gigs are pretty straightforward: show up, read the script out loud, then head home and watch people on Tumblr complain that the original Japanese voiceover was better. But, because the goal of Siri was to record basically every possible sound a human being can make, the process was the closest thing the voice-acting industry has to a death march:
“The original vocabulary took the month of July: four hours a day, five days a week. And I read phrases and sentences that were created solely to get all the sound combinations in the language. It was extremely tedious work. A couple of years ago, this company tried to put me under contract for five more years. I declined; or rather, I asked for a lot of money and they said no. It’s very, very tedious stuff — everything has to be read in the same pacing, the same timbre, the same tone. It’s challenging in its own way, and you do get a sense of accomplishment from doing it well, but at that point, I felt I’d done enough.”
Twenty hours a week may not sound like much, but extended periods of voice acting can wreck your voice, which keeps you from pursuing other work:
“It really is [tough on your vocal cords], especially because you’re not varying your voice at all. You’re reading just like this. And this. And this.” (Editor’s note: At this point Susan suddenly and effortlessly dropped into the Siri voice, and it was really creepy.) “For page upon page, hour upon hour. I took a lot of breaks; I drank a lot of water. I didn’t do much other work because it was so stressful on the voice. We did about four months of updates in 2011. For that I said, ‘Look, we’re going to cut this down to two hours a day.’ Because I didn’t want to miss other work because of it.”
Even if you can still pursue other work, talking like a patient robot doesn’t make it easy. “[Taking your emotion out of your voice for hours] doesn’t help your other areas of performance. It’s pretty monotone; it’s the opposite of what one would consider creative. Last year I took several months of voice coaching just to make sure I wasn’t stuck.”
It’s part of a process called concatenation, where vowels, consonants, syllables, diphthongs, and other aspects of speech are extracted and then reassembled by technicians to form whatever sentences they need. All of which is very fascinating but provides little comfort when you go home feeling like you’ve just recorded the audiobook for Dr. Seuss’ opium-induced fever dreams.
Siri Highlights How Voice Acting Is Screwed
At 67, Susan’s been in the voice-acting game a long time. She started out as a singer, got into radio jingles, and from a background in advertising moved into more niche work, like airline gates, phone message systems, and one of the first ATMs. And she’s noticed that, ironically, the job that got her on Letterman is indicative of the problems her industry is now facing. Apple’s hardly the only company to play fast and loose with payments. It’s common to see people not get paid for the use of their voice, she told us, particularly when it comes to non-union gigs: “One of the advantages of union work is they buy your voice for a certain period of time, and if they want to use it again they have to pay again. Now no one wants to deal with unions and paperwork; no one wants to write two checks. They just want to be done with it. All of this stuff has been devalued.”
It’s also meant mixed career prospects for Susan. Being a robot is fun, but, as so much science fiction has taught us, they yearn for more:
“I knew one of two things was going to happen. It could have been, ‘Oh, she’s Siri, let’s get her!’ Or it could have been, ‘Oh, she’s Siri. She’s everywhere already, let’s not use her.’ I’ve gotten a lot of jobs as a Siri type of voice, or a robot voice, because of this. So I’m in the process of changing some of my demos to make sure I don’t get typecast.”
“I just wasn’t sure what I wanted to do with that information. In the digital age, voice actors really have an advantage in being anonymous, unless they’re celebrities. Because no one knows what you look like, where you’re from, how old you are, and all of that stuff, they’re judging you solely on your performance and the sound of your voice. And I knew I’d be giving that up. I was in a quandary for a while; it took me a long time to decide. My husband and son both convinced me that it was such a unique opportunity and unique situation that I had to do it and see what would happen. And I’m glad I did. It’s really been a lot of fun. I’ve had the opportunity to do many things I never would have been invited to do otherwise.”
People weren’t just excited to learn she’s the person — they were excited to learn there was a person. “People just couldn’t believe there was a voice behind it. And I thought, ‘Well, where on Earth did you think it came from?’ They thought Apple just invented it somehow. Machines can manipulate a voice, but the basic sound has to come from a human. At least, so far.”
Susan’s enjoyed the fame, although she hasn’t acted on our suggestion to sneak up behind someone who’s asking Siri a question and whisper the answer in their ear. Come on, it would be fun!