Does a Sensory Input Deliver Intelligence?
A week rarely goes by without a story about the more sensational aspects of natural language hitting the headlines. It would be easy to believe that everywhere there are highly intelligent apps bowing to a user’s every spoken word. But the truth is a little different.
Don’t get me wrong, it’s very clever that authentication software can analyse your voice so you don’t have to tap in a password, or that a selfie will confirm a payment, but there is still very little cohesive understanding between everyday technology and you. Once you get into your account 15 seconds faster, it’s still just as dumb as it was before.
A banking app hasn’t understood that you’re accessing it from a foreign country and might want to know exchange rates or how much withdrawals cost from a local ATM. An app for a high street retailer doesn’t remember from the last conversation that you want a new sofa because you’re buying a house. Not to mention the missed opportunity for the company to build a relationship by starting off – Hi Andy, how’s the house move going?
The technology is there to do it today, alongside the demand from consumers; it’s the complications around delivering the intelligence that is slowing down the delivering of the dream.
Part of the reason behind this is that many companies are introducing point solutions using aspects of natural language just to solve specific issues. These can often be characterised by their use of sensory natural language inputs, such as Automated Speech Recognition (ASR), Text to Speech (TTS), gestures, face recognition and even biometrics.
The trouble with all of these inputs whether it’s spoken or written word, finger movements or voice patterns they are just a conduit for entering information. Without technology such Natural Language Interaction (NLI) there is no intelligent understanding behind them. This creates a problem, not just for the user, but for the company developing the app too.
Users are frequently frustrated by still having to use the commands set out by manufacturers and driven to distraction that simple phrases are misunderstood. Add to this the necessity of repeating every piece of information because there is no inherent memory and it’s easy to see why people perception of intelligent apps isn’t always positive.
If you were to ask a travel agent how much a flight to Bali was, and then say what about Mauritius? The chances are you’ll get the answer you were expecting. Ask a travel app and it’s more likely to reply What about Mauritius? Actually, no it wouldn’t, because that would imply a personality and a sense of humour. No the answer you would receive is Sorry, I don’t understand the question. Could you rephrase?
Building an app that meets these expectations requires the ability to understand words, concepts and sentiment. It needs to remember details, to learn from interactions a user’s preferences, to take information from other sources and collate everything together. And most importantly, it needs to return the correct response immediately.
In the past, achieving just a fraction of this took an army of computational linguists many months of development effort. Teneo, Artificial Solutions’ NLI platform, with its built-in artificial intelligent capabilities has changed all this.
For example, most businesses literally have to think through the hundreds of ways a customer might ask a question, and then code each one into its natural language interface. Teneo, which has vast language resources to draw meaning from sensory inputs such as voice, requires only two sample questions to each answer to deuces the rest itself.
Other capabilities such as memory, machine learning, implicit personalisation and persistence across devices, all come together within Teneo’s NLI engine to be interpreted, analysed and reasoned out, in much the same way a human brain works.
Whether it’s the IoT, Digital Employees or a propriety interface, regardless of the sensory input, Teneo allows enterprises to deliver a sophisticated intelligent experience from day one. Enabling technology to revolve around your life, not the other way around.
The Impact of Conversational AI on Connected Consumer Goods and the Internet of Things White Paper
This white paper reviews the art of the possible with conversational AI and reveals how tomorrow’s technology can be achieved today.