A Generation Later, and Speech Technology Finds Its Place in Customer Service

In its earliest iterations, speech recognition technology was designed to do simply that: recognize the spoken word and translate it into written text. IBM (News - Alert) computers began showing up in the 1990s that allowed users to dictate text. The results weren’t very precise: it didn’t cope well with people with certain accents, for example, or background noises.

Several generations later, and speech recognition has morphed into natural language processing (NLP). This means technologies can not only recognize speech, but also understand the meaning behind the words.

“You need solutions that can understand intent and drive automated responses,” Daniel Ziv, vice president of speech and text analytics and global product strategy at Verint (News - Alert), told Destination CRM. “To figure out what those questions are, you need solutions like speech and text analytics that have NLU embedded in them to mine the incoming conversations in unstructured voice or text.”

It’s not enough for a solution to simply recognize words: it needs to understand what the speaker is really looking for. Natural language can also engage in speaker triage: that is, separating routine inquiries from urgent issues, and directing those urgent issues to a human agent first.

Speech Technology Isn’t Replacing Humans

When automated contact center technologies such as natural language processing first hit the market, the headlines were predictable: “Human agents may be replaced by robots.” While NLP may have turned human receptionists into an endangered species – directing calls is a pretty basic job that can be easily automated – it’s unlikely to replace humans anytime soon.

“Technology marches relentlessly forward, and it would be foolish to argue otherwise, but some things remain fundamental, and people-to-people communication will continue to be one of them,” wrote Ron Miller for TechCrunch. “Just because the tech is available, doesn’t mean it’s always going to be the best option in every situation.”

Even as speech technology gets better at divining a caller’s intent, it still won’t be able to provide the human touch that customers expect. This includes understanding complex problems, collaborating with other humans, and offering sympathy. Smart companies are using NLP in the contact centers to help direct calls and other contacts, triage them and sort them into proper handling order. It can also help at the back end, handing a live customer call over to a routine information search, for example.

The best approach to natural language processing is to think of it as a high-tech personal assistant to human agents, making their jobs easier, speeding up the time to call completion, and complementing the human touch. 

Edited by Luke Bellos