Secret Agents Doubling as Virtual Assistants
Posted on August 14, 2015
Humans continue to be more intelligent than machines
Speech recognition is continually improving. Note the functionality that Google Now and Apple Siri provide from devices that are notably difficult to interact with using text. However, has it really changed much in the contact center? Sure, a few large corporations have implemented natural language understanding or “conversational” IVR apps using traditional speech recognition. The seven-digit cost, covering the cost of the software and ongoing very expensive tuning behind these apps precludes anyone but the largest organizations from deploying these solutions.
However, even for well-designed apps, the current state of ASR still limits the ability to expand self-service past very simple apps that are limited to only a few caller inputs. The danger of even moderately complex speech-enabled apps is the dreaded “I’m sorry, I didn’t understand. Could you please repeat that?” which usually drives the caller to press zero or repeat “Operator, operator!”
So should we give up all hope for speech recognition in the contact center unless you are a massive organization with plenty of cash? Absolutely not! Human Assisted UnderstandingTM, a technology developed by Interactions LLC, is an extremely potent and cost effective means to improve caller intent and expand speech-enabled self-service. Variations of it have been available for over ten years, but the market uptake was not high until two or three years ago. Adaptive-UnderstandingTM works by understanding an utterance’s confidence level, and will send an utterance that fails to meet the ASR’s ability to understand the utterance to a human along with a screen pop of most likely matches.
Over the history of the HAUTM and similar offerings, several companies have used different role titles, but in business conversations, I tend to use “secret agent.” The secret agent matches the utterance to the correct match and the application progresses without the caller knowing that a human was working behind the scene for several seconds. The human ear is not perfect, but it still is much better at disambiguation of speech than current speech recognition technology. The NLU apps mentioned earlier require significant amounts of tuning to ensure the grammars contain enough words that will optimistically match almost all of callers’ possible utterances, but not too many words to allow misidentification of non-targeted words as the correct utterance. Moreover, the tuning to keep the voices of two humans, or even just the caller’s utterances on track (to the IVR: “Tomorrow” immediately followed to the child/pet, “I said stop it!”) is a complex balancing act. It falls somewhere between science and art, and does not come cheap.
In contrast, HAUTM is very accurate immediately upon deployment and when combined with additional Adaptive-UnderstandingTM technology, it quickly and cost effectively tunes the ASR grammars. To be transparent, I have been a major proponent of the capabilities of HAUTM since it first hit the market. Unveil Technologies first offered similar technology, but Microsoft acquired them in 2005 and pulled the offering from the market. At least two others attempted to replicate the technology, but for various reasons, the market did not adopt it. Several years ago, Interactions LLC began offering HAUTM, just as the market awoke to the understanding of what HAUTM can do for them at a cost effective price point. Late last year, Interactions and AT&T completed a transaction where Interactions took possession of AT&T’s Watson natural language understanding technologies, along with several other assets. This acquisition bodes well for continuous improvement in use of Adaptive-UnderstandingTM for tuning, as well as extension into additional customer contact channels.
Extremely successful hosted, speech-enabled virtual assistants with high ROIs are now within the reach of organizations big and small. It is great news for callers, and organizations’ service and sales teams alike. Imagine hospitality reservations or the scheduling/changing/cancellation of home service appointments, perhaps full enrollment into health insurance programs; all accomplished 100% via voice virtual agents. To spark your imagination further, what if these voice channel virtual agents perform so well, they have higher NPS and customer satisfaction scores than when your own employees handle those calls? I have been involved with implementation of the technology, and it truly is here, now.
For those of us old enough to remember, this is where Don Adams as secret agent Maxwell Smart in the Get Smart! TV series would likely remark: “Would you believe…..the old IVRs transforming into virtual assistants trick?”