Speech Technology Magazine

 

The Evolution of Virtual Assistants

Siri may be at the forefront, but she's far from on her own.
Posted May 10, 2013
Page1 of 1
Bookmark and Share

I love talking about virtual assistants. I first wrote about them in 1995 under the moniker "personal telephony," when a plethora of companies--up to 40 or more--introduced the first telephony-based, speech-driven assistants. Those in the industry back then might remember HeyAnita!, Webley, General Magic, or Wildfire, products that enabled people to use voice to check stock prices, sports scores, and the weather, or control telephony and office productivity functions, like updating calendars and making conference calls.

We have come a long way since then. Just ask Siri, the virtual assistant at the forefront of making speech assistants a very consumer-facing solution. Siri, in particular, because of the popularity of the iPhone and the marketing dollars of Apple, really put a face to a voice when it came to using speech as a personal concierge. She enabled consumers to access all types of information and perform a great deal of productivity actions to make it simple to get things done from anywhere.

But Siri isn't the only girl in town. Sophia, Lexee, Nina, and others are consumer-facing from a different perspective--serving as the voice user interface (VUI) to mobile customer care applications. Last year was when these types of applications really took off. In all, more than a dozen mainline contact center vendors came out with mobile customer care solutions aimed at bridging the gap between apps on a phone and phone applications that provide customer care the way customers would expect it from live agent assistance. Among those that launched solutions were Interactive Intelligence, Genesys, Cisco, NICE, Angel, Nuance, Voxeo, and XTone.

While a true VUI was central to only a few, speech technologies proliferated throughout, with text-to-speech, voice dialing, speech-to-text, and voice authentication. Developers of these products worked hard to use the unique properties of mobile devices to enhance the customer experience, by providing multimodal and multichannel capabilities, voice, tap or type interfaces, geolocation capabilities, and pictures and video. In 2013, expect to see more VUIs and more uses of speech added to these first-generation products.

The evolution of speech assistants continues on a different front as well. While mobility took center stage in 2012, other virtual assistants--"agents" in the contact center--marched on. There are numerous vendors whose charter it was to create speech-enabled agents to assist customers in lieu of live agents. Companies such as Creative Virtual, VPI Corporation, and SmartAction, for example, have created intelligent virtual assistants who interact with customers over the phone, helping to alleviate much of the traffic coming into the contact center.

Where we are headed now is even broader. Virtual contact center agents have been around for a while now, but have evolved to handle more than just voice. In addition to those classic agents, and now mobile assistants, other channels are coming into play--in particular chat. Creative Virtual's V-Person technology can be deployed as voice, mobile, or Web agents, accessing an integrated central knowledge platform for cross-channel integration and handling of customer interactions. [24]/7 Online uses predictive analytics, real-time decisioning, and rich multimedia widgets for guided self-service for complex tasks, making for a more intuitive self-service experience.

Creating a great VUI, particularly one that can mimic an agent rather than just employ directed dialogue in an IVR application, is no easy feat. But we have come a long way from when we did just that. We have improved the natural language interfaces, predictive analytics, artificial intelligence, cross-channel analytics, and knowledge management databases to make self-service without live agents an attractive option for companies. With some of the newer technologies coming into play, such as self-learning capabilities of these "agents," along with predictive analytics and big data, virtual agents across all channels can only get better.

Nancy Jamison is a principal analyst for contact centers at Frost & Sullivan. She can be reached at nancy.jamison@frost.com, or follow her on Twitter @NancyJami.

Page1 of 1
Learn more about the companies mentioned in this article in the Speech Technology Buyer's Guide:
Learn more about the companies mentioned in this article in the Vertical Markets Guide: