Copyright ©2006 NPR. For personal, noncommercial use only. See Terms of Use. For other uses, prior permission required.

MADELEINE BRAND, host:

This is DAY TO DAY from NPR News, I'm Madeleine Brand.

ALEX CHADWICK, host:

I'm Alex Chadwick.

You call a customer service line, dear listeners, do you get help or do you just get aggravated? You try to clear up a bill with an automated phone system, you get a million questions from a robot and then you get put on hold. Well, maybe what you need is help from artificial intelligence. DAY TO DAY Technology Contributor Xeni Jardin is here with more.

XENI JARDIN: Okay so you call up your bank about a late fee and you might get this response.

Unidentified Man #1: Our records show that your payment was due on the 15th and was received on the 17th.

JARDIN: But of course you, the customer, disagree.

Unidentified Man #2: That is bizarre I did a electronic payment on the 13th.

JARDIN: And then you get some gobbledygook explanation.

Unidentified Man #1: Well, you know, many aren't aware that electronic payments aren't instant. And in fact it could take up to five days...

Unidentified Man #2: What are you kidding me? Your bank takes five days to process an electronic payment?

JARDIN: And so you get angry and play your final card.

Unidentified Man #2: Listen to me very carefully, you'll remove the entire $40 or I will pay off my entire balance and close the account this month.

JARDIN: This is a corporate demo, it's bought to you by NICE Systems, a technology company that has developed software that records calls and listens for emotional signals that the call is going badly. It could be valuable to Fortune 500 companies because it would help them spot problems in customer support and improve their service.

Ms. ROBIN SCHAFFER (NICE Systems): They can flag and look for calls where customers are saying late fee or cancellation, or I hate you or other phrases.

JARDIN: Robin Schaffer from NICE systems says the technology isn't just listening for troublesome words. It wants to know how you feel.

So what it does is it essentially takes a snapshot of the voice - and that includes many, many parameters not just volume which you know many of us think it's the most significant - in fact its not. It's pitch, it's intensity, intonation, speed, it's a lot of different factors.

JARDIN: The software analysis all of those human factors and processes them with algorithms. That information generally won't help right there, during that call, but Schaffer says it can be used to answer important questions companies may have.

Ms. SCHAFFER: What kinds of situations - what kind of topics are bringing up this kind of level of emotion. And it could go down to the soft skills or particular agents that seem to have patterns of customers becoming more emotional. Or it could come to broad business issues, that - where you can notice a pattern between heighten emotions and a particular product or service, or geographic area, or some other very important business component.

JARDIN: Phil Tschudy is from CUNA Mutual a financial services group in Madison, Wisconsin that processes thousands of calls a day.

He says his company has been trying out the software for about a year as a feedback tool.

Mr. PHIL TSCHUDY (CUNA Mutual): This is an opportunity for us to go back and make sure that we're looking for trends that can be addressed, that we can help our call center representatives improve the services they deliver.

And I think ultimately consumers want that, and demands are higher than they have every been. And so this is great tool for us to really understand what are customers want and then deliver on that - on those expectations.

JARDIN: This idea sounds great, but there is some problems. Big companies want to cut costs, so more phone support has been outsourced to call centers overseas. But the issue is that English may not be a first language for most of the reps answering phones. And if a communication gap grows wider another automated solution probably won't go far enough to comfort the outraged and unhappy customer. For NPR News, I'm Xeni Jardin.

(Soundbite of music)

Copyright © 2006 NPR. All rights reserved. No quotes from the materials contained herein may be used in any media without attribution to NPR. This transcript is provided for personal, noncommercial use only, pursuant to our Terms of Use. Any other use requires NPR's prior permission. Visit our permissions page for further information.

NPR transcripts are created on a rush deadline by a contractor for NPR, and accuracy and availability may vary. This text may not be in its final form and may be updated or revised in the future. Please be aware that the authoritative record of NPR's programming is the audio.

Comments

 

Please keep your community civil. All comments must follow the NPR.org Community rules and terms of use, and will be moderated prior to posting. NPR reserves the right to use the comments we receive, in whole or in part, and to use the commenter's name and location, in any medium. See also the Terms of Use, Privacy Policy and Community FAQ.