IVR – technology for a great compromise – PART 2

Part 1 of this article was posted immediately before the Part 2. I recommend to read it first.

If you will use this text for publishing or academic purposes, be so kind to cite the author and source: Alen Gojceta, Banka, 05/2002. Thank you!

The Future of IVR

Except the growth of the number and size of call centers and the need to reduce their labor costs, the IVR market growth forecasts rely on new technologies that enable easier and more natural interaction comparing to the one based on tone dialing and pre-recorded speech sequences.

Here we primarily emphasize Natural Language Speech Recognition (NLSR), Voice Recognition (VR) and Text-to-Speech (TTS) technologies. Growing trends are as well driven by the new approach to using IVR technology, so called voice portals.

When it comes to advanced IVR technology, the future has already begun. Many solutions already exist. Due to the complex technology, there are a few suppliers of functional NLSR, TTS and VR software products in the world. To support these technologies, the leading providers of IVR systems usually integrate technological solutions by niche software vendors such as Nuance and SpeechWorks.

NLSR technology supports giving instructions to IVR by voice, using natural language. In the case of well-designed system, an order will be given in spoken like: “I want to make 550$ transfer from my savings to my current account”. The machine will perform the operation without employee interference. NLSR makes possible what was hard to imagine, and in some cases, it is inevitable replacement to the traditional service management based on tone dialing.

TTS technology enables translation of any digitally stored text to speech. In this way it is possible to dynamically leave a message to a caller. For example, a bank client contacts the call center to check the status of his account or other routine interaction. After identification of the client, the IVR automatically communicates a personalized message that was entered into the system in form of text by a bank employee in the loan approval department: “Dear sir (Jones) we are still waiting for your mortgage estimation to close your credit claim. Please contact mrs. Patty on 123456. Thank you.”.

VR technology is based on voice recognition algorithms that rely on original features of each individual’s voice. VR is one among many increasingly popular biometric identification methods, based on specific characteristics that are unique for each of us, such as fingerprint, eye pattern, and even the DNA structure.

All this technologically demanding solutions are based on complex mathematical algorithms and artificial intelligence technologies, such as neural networks. Important role in its success have the ever more affordable processor power and storage capacity.

The companies providing these technologies have operated at loss for years, funded by capital from different sources. We are witnessing a market capitalization of the few that survived. New technologies are constantly making management of IVR services easier, transforming in this way the IVR platform from a system of compromises to a system of customer desires.

Voice Portals

Voice portals are single access points that allow a caller to retrieve different types of information or manage personal communications services, by voice using the telephone service. Voice portals integrate Natural Language Speech Recognition (NLSR), Voice Recognition (VR) and Text-to-Speech (TTS) technologies with the communication infrastructure and application expertise. Simply put, voice portals are IVR systems with extended functionality resulting in increased ease of use and a better flow of information, thanks to the use of new technologies.

The concept of voice portals is very similar to the concept of web portals, so the supported types of services are very similar to those on the Internet. Most of these services will be focused on content, communication, voice commerce (as opposed to online store) and remote access to business resources.

The advantage of voice comparing to web portals, is in the handy nature of telephone communication. Consequently, the information that the user “pulls” by a phone will be the one that is urgent, updated and time sensitive.

As in the case of web portals, business organization behind the voice portal is a labor-intensive and complex. Therefore, their maintenance is left to specialized organizations that provide such services on the market. In line with this need, recently a new category of service providers appeared – Voice Application Service Providers.

Today (2002) we can identify a large number of application cases using the advanced IVR technology in various industries such as booking systems in tourism and passenger transport, access to financial data and performing transactions in banking, retail and catalog sales, accessing information of the government administration, CRM applications at mobile service providers…

As already mentioned, the Croatian market of “traditional” IVR systems is well developed. We will still have a lot to wait for new technologies like NLSR or TTS due to high cost of development of new languages patterns. But this is the fate of all small “non English speaking” markets.

END