Natural language processing has come a long way in the last few years, as digital assistants such as Apple Inc.’s Siri being able to listen to human speech and understand exactly what was said. Siri also can act on those words, for example by providing answers to people’s questions or requests for information.

But true understanding of human speech involves a lot more than simply comprehending people’s words, because the manner in which people speak also plays a key role in how others will interpret them. With this in mind, IBM Corp. today added a new feature to its Watson service called “Tone Analyzer for Customer Engagement.” It can detect the tone of people’s voices so that computers and humans can respond more appropriately to what’s being said, figuring out emotions such as frustration, satisfaction, excitement, politeness, impoliteness, sadness and sympathy.

Tone Analyzer was first trained on customer support conversations on Twitter, and IBM reckons companies can use it to monitor their own conversations with customers with a view to improving the level of service they offer. The technology can be used by both human call center operators and also by chatbots, allowing them to provide more appropriate responses to frustrated or angry customers on the phone. IBM said this is important because communication often makes all the difference between making a sale or losing business.

The technology should prove useful in certain situations, for example if a call center operator in India cannot quite grasp a caller’s accent, or if the quality of the call itself is bad. Tone Analyzer can help operators to interpret a caller’s emotions, so they’ll know the person they’re speaking to is angry or frustrated even if they can’t quite catch everything that’s said.

In addition, the service can help chatbots to know when a customer is upset by reading the messages they type. This allows the chatbot to come up with a more sympathetic response that will hopefully calm the person down.

IBM said that the technology has already been put to good use by a number of early adopters, including KNOLSKAPE Labs, which is a talent transformation company based in India and Singapore. KNOLSKAPE has been using Tone Analyzer for several months, testing its employees in simulated customer service scenarios to evaluate the appropriateness of their responses to questions. Tone Analyzer provides feedback to the employees in real-time, showing areas in which they’re doing well and also where they need to improve.

“The quality of communication between the company and the users is a top priority for businesses where the majority of communication happens with customer service representatives in English,” said Chaithanya Yambari, director at KNOLSKAPE Labs. “Tone Analyzer for Customer Engagement gives valuable metrics which help in assessing the quality of this communication to train customer service agents to be consistently polite.”

IBM said Tone Analyzer for Customer Engagement is generally available now, and is currently offering a free demo for customers who wants to try it out.

Image: OpenClipart-Vectors/Pixabay