Normal speech delivers a Black Cab

Implementations of interactive voice response (IVR) systems with natural language speech recognition (NLSR) are growing in number.

The latest recruit to the technology is one of Australia's largest taxi operations, Melbourne-based Black Cabs Combined Limited.

Scheduled to go live in late January 2001, the company is implementing an IVR customer cab ordering system with NLSR capability at its call centre.

Black Cabs dispatches bookings to nearly 1500 taxis within Melbourne's metropolitan area and surrounding suburbs. The call centre processes more than 6.5 million customer calls a year, excluding taxi 'flag' customers.

Garry Bradd, chief executive officer for Black Cabs Combined, said he does not expect any integration issues during the implementation. He said the system is capable of linking straight into the company's Nortel PABX and the computer dispatch system used by taxi drivers.

He expected the system, developed by Australian call centre solutions provider Premier Technologies, would increase the centre's call handling capability by 10 per cent during peak periods.

The automated system will initially handle 60 calls simultaneously. Bradd said the system will be offered only to customers that the system "recognised", such a those whose address or phone number was previously filed. The system will verify and reconcile caller details and requirements by asking a series of questions, and process customer's answers regardless of accent.

"If the system doesn't recognise the customer in two goes, the call will be switched to an operator," Bradd said.

"Within one month we expect to achieve 96 per cent accuracy for first efforts; currently we believe our telephonists offer 88 per cent accuracy.

Bradd said the call centre would also be used in other areas of the company, such as marketing.

• Bookmark this page
• Share this article
  •  
  • facebook
  • slashdot
  • digg
  • Reddit
  • stumbleupon
  • linkedin
  • twitter
• Got more on this story? Email Computerworld
• Follow Computerworld on twitter

• Share
  Share this article

  •  google+
  • facebookfacebook
  • slashdotslashdot
  • diggdigg
  • RedditReddit
  • stumbleuponstumbleupon
  • linkedinlinkedin
  • twittertwitter
• print
• email
• Bookmark