I am in the business of building robots that not only say things using a human sounding voice but converse the way human beings converse with each other (or at least try to converse). This means you talk to them, they talk back to you, then you respond to them, and so on, in a back-and-forth interaction. In other words, I am in the business of building what are called voice-based, conversational robots, or “voicebots” for short.

Building such voicebots gets you thinking deeply about matters like language, conversational rules, norms, politeness strategies — things that make us human beings human. But in the quest to build robots that emulate quintessentially human behavior, I got to learn quite a bit about not only language, social norms, and conventions, but about myself, with some concrete lessons on how to become a better human being.

I share five of these lessons here.

Speak Less
It continues to happen to this day, almost seven years since Amazon Alexa and Google Assistant were born: the voicebot often talks way more than it should. Many (probably most) Amazon Alexa skills and Google Assistant actions will start their interaction by reading to you a carefully written out paragraph, or several such paragraphs, as if what they were telling you was very important. It never is. Which made me realize this about myself: It is almost never the case that what I have to say to anyone at any point is very important. It’s probably barely worth listening to, let alone remembering. With that in mind, I have committed to keeping what I say to as much of a minimum without being cryptic. (By the way, this is hard – I have a long way to go.)

Lesson: ‘Nuff said.

Aim to Communicate
Having engaged with badly designed voicebots, I have also come to learn (or at least to cement firmly and keep top of mind) some basic, fundamental communication best practices. Like this one: place the focus of your intent last in a sequence. If your intent is to suggest to someone what to eat, instead of saying, “Eat an apple, not a kiwi,” say, “Don’t eat a kiwi. Eat an apple.” If your intent is to suggest to someone that they should avoid kiwis, instead of saying, “Don’t eat a kiwi. Eat an apple,” say, “Eat an apple, not a kiwi.” Or this one: make sure that you never say something that leaves your listener wondering: was that a multiple choice or a yes/no question?

Takeaway: Focus on communicating intentionally.

Say/Do the Right Thing at the Right Time
Scenario One: I have my iPhone hooked up to my smart speaker in my veranda, listening to a podcast as I sunbathe. I am hungry at some point and need to get myself a snack. I get up, pick up my phone and go to the kitchen, where another smart speaker, already paired to my iPhone, sits. Like magic, the iPhone switches to my kitchen speaker and I continue to listen to my podcast. I smile and nod in appreciation and tip my cap to the designers for being so thoughtful.

Scenario Two: I have my iPhone hooked up to my smart speaker in my office downstairs, listening to music as I work. It is very early in the morning and the rest of the family is still sleeping. I am hungry at some point and need to get myself a snack. I get up, pick up my phone, and go to the kitchen, where another smart speaker, already paired to my iPhone, sits. Like magic, the iPhone switches to my kitchen speaker and music starts blaring. I unplug the smart speaker, cursing at the designers for being so thoughtless.

Bottom line: Behavior that delights under certain circumstances is detestable under other circumstances. Understand the context.

Human Beings Become Blasé Very Quickly
In 2016, on a visit to Algeria, my country of origin, I took with me my tall, slick, and slender black Amazon Echo. The Echo was a novelty at the time, even in the U.S., and almost unheard of elsewhere in the world. No one in my family had heard of it, let alone seen or touched one. And so, I set it up and began interacting with it, to the “oohs” and “aahs” and astonishment of those to whom I showcased this magical device. But fifteen minutes later, however, I started hearing things like, “Alexa, shut up!” or “Alexa, you are dumb!” ‘Well, that was quick,’ I distinctly remember myself thinking. I was surprised to discover that I was delighted that the Echo, sorcerous as it may have seemed to me, since I understood how much it took to deliver such a technological marvel, was treated by non-experts as a piece of hardware that could do things for them, and not much more.

Upshot: Never rely on novelty, or engineering virtuosity, or bells and whistles, to guide your design goals (or your sales strategy). They wear off quickly. Instead, focus on delivering (and selling) lasting value.

Hold No Grudges
This is a true story that makes people chuckle whenever I tell it to them. Once I was driving from my home in McLean, Virginia, to somewhere in downtown Washington, DC that I had never been to before, and so, as usual when I am going to a place I don’t know the way to, I fired up Google Maps and turned on directions. But since I have gone to DC innumerable times, the directions that I needed were really for the last few miles of the trip. Between the start of my drive until then, I knew what routes to take, the right shortcuts, and what to avoid given the time of day and day of the week. And so, listening to some music, I kept ignoring the assistant’s instructions, since they were not optimal. “What a dumb machine,” I kept thinking, taking pleasure in completely ignoring the assistant as it tried, frantically it seemed, to get me back on “the right path.” Then, suddenly, the assistant went silent. For miles and miles, it was just jazz music. Not a peep from the assistant. Then, as I neared my destination, the assistant was back on. Earnestly giving me directions that I now needed, cheerfully, with enthusiasm, with focus and care, not a hint of resentment or sulking in its voice. I distinctly remember feeling impressed by this robust robot that had stayed focused on the task rather than dwell on a slight or my misbehavior. (I also felt a bit ashamed of myself for mistreating this wise bot.)

Moral: You come across as strong and admirable if you keep your ego in check and instead focus on the task at hand. Pride, after all, is the deadliest of sins, and it’s a good thing that our voicebots have (and hopefully will continue to have) none of it.

Dr. Ahmed Bouzid, is CEO of Witlingo, a McLean, Virginia, based startup that builds products and solutions that enable brands to engage with their clients and prospects using voice, audio, and conversational AI. Prior to Witlingo, Dr. Bouzid was Head of Alexa’s Smart Home Product at Amazon and VP of Product and Innovation at Angel.com. Dr. Bouzid holds 12 patents in the Speech Recognition and Natural Language Processing field and was recognized as a “Speech Luminary” by Speech Technology Magazine and as one of the Top 11 Speech Technologists by Voicebot.ai. He is also an Open Voice Network Ambassador, heading their Social Audio initiative, and author at Opus Research. Some of his articles and media appearances can be found here and here.

Categories: Conversational Intelligence, Intelligent Assistants, Articles