Speech Technology Magazine

Examining speech application errors reveals several more likely causes.

Taking advantage of scalability and support puts the focus on customer experiences.

With new markets come new uses for the technology.

Harking back to a simpler day when phones were less advanced.

Enterprises must find the right balance between live service and self-service automation

How to present geographic locations to Spanish-speaking users of automated systems in the U.S.

How a larger taxonomy of errors can inspire change.

Don't harp on errors that can be resolved later.

Reprompts should correspond to the type of error being generated.

Giving callers a sense of power during a call increases the chances for task completion

For speech to move into the mainstream, VUI designers need to jump in early.

VUI testing should involve surveys, call recordings, and call logs together.

The telecommunications industry is undergoing a major change from proprietary hardware-based systems to software-based systems built on open standards. Standards such as VoiceXML, CCXML, SIP, Service-Oriented Architecture, and other Internet technologies are radically changing how contact center applications are built and deployed. But these enabling technologies are only the means to an end. The real goals of applying new technologies to the contact center remain improving the customer's experience.

Keep it simple. Such a straightforward goal, but one that is so often sacrificed in the search for engineering efficiency. My latest encounter with this penchant for complexity came in the form of the over-engineered elevator system at the Marriott Marquis in Times Square, host to SpeechTEK 2005.

James Wimberley explains, "A speech recognition module tuned too finely to native speakers would be frustrating for learners; one that trained itself to accept their mistakes uncritically would miss an important learning opportunity."

Guy Alon, marketing manager of NSC, explains how the higher education market can be a "huge potential of using speech recognition."

Gary Tjaden, founder and CEO of the Cocomo ID Company, explains that "there is a much more cost-effective approach to speech-audio publishing than the current digitized speech approach, which is to use a text-to-speech (TTS) engine embedded in a PDA or cell phone to speak general information downloaded as small text files.

Terry Thompson, senior computer specialist at the University of Washington, explains, "If industry players embrace universal design, all users - with and without disabilities - will finally be able to participate in our technology-enhanced world."

Bill Meisel, president of TMA Associates (www.tmaa.com), explains why "The ScanSoft-Nuance merger, with ScanSoft agreeing to purchase the outstanding shares of Nuance and use the name Nuance for the merged company, will send shock waves through the companies' partner networks and end-user customers."

Michael Chavez, vice president of client services at ClickFox, explains that "by leveraging a strategic combination of customer behavior intelligence, customer service interviews and surveys, organizations can reduce customer frustration with IVR systems, which will result in drastic savings, while also improving customer satisfaction."

According to Sumit Badal, senior staff software engineer with Nortel, "the Grid is coming - and so is unconstrained speech recognition."

Travis White, vice president of marketing and product management at Datria, explains that predictable service should be a company's number one priority above all else.

Peter Leppik - pleppik@vocalabs.com - takes a look at improving customer service.

Gone are the days of big deals meaning big discounts for IVR ports. The new definition of success is how many calls can be resolved and how fast the next app goes into production.

The term "conversational" has been over generalized to refer to only one kind of conversation. His article explores in depth how designers can find a balance between maintaining the conversational flavor of speech while designing for more frequent use.

When one of the nation's leading telecommunications companies introduced an automated speech application to provide independent verification of its new customers - a practice required by the FCC - it approached the "go-live date" with some trepidation. On that day, the automated application would replace live agents as an initial point of contact for new customers - a critical point in the customer relationship. How would customers respond?

Speech application developers have long understood the importance of good voice user-interface design. The telephone is the most intimate form of communication most consumer product and service companies have with their customers. Unlike advertising, or even the Web, it's the contact medium where customers are the most engaged and prone to form impressions, good or bad, of a company.

Text-to-speech (TTS) technology leaves many application designers at a crossroad. Although TTS technology is deployable - and it frequently sounds great - it still makes mistakes. What's more, it makes some mistakes that seem painfully obvious to us.

Several billion times each year consumers call over wireline and wireless networks and pay a transaction fee to get a telephone number. Directory Assitance (DA) Service Providers in North America spend hundreds of millions of dollars to offer this service, with most of the money paying the cost of the operators.