What and Why – Conversational Artificial Intelligence

Communication through Artificial Intelligence

Communication is key they say, or at least I was told so when I was younger. However, the older I got, the more proof I saw to this age-old saying. We more often than not end up misunderstanding the other that helped create opportunities that allowed ambiguity to prosper.

Communication is important, and it gets a lot more important when we’re talking business. We’re taught back in business 101 that effective and efficient communication is imperative for a business owner if you’re looking to run a successful business. From getting our points across in a manner that allows every party to fully understand what’s being said; in business, good communication is the difference between closing a deal or missing out on a great business opportunity.

Conversations Arm Communication

Think of communication as a mix of concrete and steel, the kind we use to erect large obnoxious concrete jungles fuelled by capitalism, coal, and cash. Conversations exist within communication, as one of the subsets of communication, while conversations are looked at as that bit of reinforced steel within that concrete mix. Similar to how reinforced steel makes your structures a lot more solid, effective and strong; the same applies to an effective conversation through communication, and AI-powered conversations do exactly this for us, whether in business or our personal lives.

With conversational AI, we’re looking at what has increasingly become the bridge between humanity and computing.

Think about it. Your ability to converse with another human is only as effective as your ability to be understood and articulate your thoughts into verbatim.

It’s a no brainer. Similarly, with the advent of concepts such as deep learning, artificial intelligence, and neural networks; we’ve got machines that can communicate with humans on a level that was totally unperceived as close as 20 years ago!

How Did Conversational AI start?

Let’s try and roll the reel back on this one. Based on Tom Gruber’s recent talk back in, January at Beacon street with John Markoff, a NY Times writer; I was able to better look at and understand when the inception of the concept i.e. conversational artificial intelligence took place, and why.

Looking to fill that gap where we needed something futuristic enough while as innovators we pushed the boundaries of what we could do with today’s technology had something to do with it. This and the fact that just like fashion, technology has and is always flirting with and maintains a rather cordial but yet personal relationship with sci-fi, and as we all know science fiction predates almost all creations. It almost acts as this predictor of things that are going to be in our systems and a part of our systems even before we’ve got that product or service rolled out.

For those of you’ll that aren’t aware, Gruber co-founded Sisi Inc., back in 2007, Siri Inc., created the Siri intelligent personal assistant. Siri was eventually acquired by tech giant Apple back in 2010 – fast-forward almost a decade later and Siri is an integral part of Apples’ iOS.

If you’re looking to jump down a rabbit hole of conversational personal assistants and how tech giants planted the idea in the people’s minds all of over 30 years back, and when John Sculley was CEO of Apple back 87, you should check out Knowledge Navigator.

Personalized Digital Assistants

Gruber helps understand why it’s not too distant in the future when all of us will have our own personalized digital assistants, this coupled with the fact that AI has already become as ubiquitous in our lives makes for a very believable case.

Back in 2008, there was a convergence of technologies that were then available for the first time in history. Concepts like cloud computing too birth, the iPhone, the app store. Essentially the idea of a microphone and a state of the art supercomputer in your hand that’s connected toa huge cloud resource had never been there prior to 08’. Gruber calls these parts along the timeline as important pieces of the puzzle, and we can understand why.

When we jump into the world of speech recognition, it requires large amounts of data in order to perform at an optimum level. It’s a data-driven process. What Siri predicted was that there would be enough data available after Siri went mainstream for a while, post which it could bootstrap itself into effectively a human-level speech recognition. This right here is what Siri is as of today. Years and years of data have been understood and assimilated, where machines are prepped and on-call to answer most of our questions and make our lives a lot easier. It ties back to the start of our article, clear and precise conversations translate to effective communication.

Let’s break this down further to get a better understanding of what Gruber is talking about and maybe we can see how the same applies to our Google Assistants and Alexas.

As of 08’ it was just speech, but all that changed over the course of time.

Contracts of a Personal Assistant

Today, assistants understand the intent and have the ability to act on this intent. This is what experts call the ‘contract’ of an assistant. We say something, once this happens, the words are transcribed (this parts called speech recognition) once this is done, NLU or natural language understanding makes sense of it and then that has to be also matched to a computer knowing to act on your desired intent.

“Human language is compositional, so getting a sequence of words to be recognized is only as good as that sequence of words occurring in some statistical population that you can then build a model on and predict from.” – Tom Gruber

About interface

Currently engaged with several enterprises in the Americas, Europe-Middle East-Africa (EMEA), and Asia-Pacific region, interface’s Intelligent Virtual Assistants or IVAs make every digital channel of an enterprise intelligent. With rich IVAs, an enterprise can leapfrog customer & employee experience to voice-first natural language interface. For more information, check out interface.ai