Speech Technology Magazine

 

Blurring the Lines: Merging Speech and VoIP

Late-at-night homeseekers in the Fredricksburg area of Virginia pass a home with a Coldwell Banker Elite "for sale" sign. They don't have to wait until the next day to get more details. If they call the real estate agency's phone number and enter the code listed on the sign, they will hear information about the particular listing spoken using text-to-speech synthesis. …
By Phillip Britt - Posted Jan 1, 2006
Page1 of 1
Bookmark and Share

Late-at-night homeseekers in the Fredricksburg area of Virginia pass a home with a Coldwell Banker Elite "for sale" sign. They don't have to wait until the next day to get more details. If they call the real estate agency's phone number and enter the code listed on the sign, they will hear information about the particular listing spoken using text-to-speech synthesis.

If the details further pique their interest, they can leave a message for the proper agent simply by dialing a three-digit extension. If they call any of CB Elite's three locations during business hours, simply dialing that three-digit extension connects potential home buyers to the agent's desk, regardless of which location is his or her "home" office.

This is an example of one of the most potent combinations of two of today's fastest growing telecommunications technologies:  speech synthesis and Voice over Internet Protocol (VoIP) delivery.

VoIP enables a person to make telephone calls using a computer network or over a data network like the Internet. VoIP converts the voice signal from the telephone into a digital signal that travels over the Internet then converts it back to an analog telephone signal at the other end. So the person at the other end doesn't need to have VoIP capability.

Because VoIP is digital, it can offer computer-enabled features and services that are not available with a traditional phone, such as speech and PBX capabilities. The user installs the application only once on the network. With a traditional phone system, speech and other applications need to be installed at each location.

"Most of the benefit in using VoIP in combination with a speech application is the ability to centrally install and administer the application," said Elka Popova, research manager, IP communications and enterprise solutions for Frost & Sullivan, San Antonio, Tex. "VoIP enables you to dynamically administer speech and other applications. You can allocate bandwidth to where you need it and operate a more intelligent network."

According to Ken Breen, Coldwell Banker Elite president, CB Elite has benefited from the marriage of the speech and VoIP technologies to provide 24/7 service, branded "Talking House" to list (sell) customers and easy access to information for potential buyers. CB Elite has saved more than $1,000 per month on long-distance charges and uses the system's tracking information for follow-up calls to potential customers.

The lower long distance costs are one of the major draws of VoIP for many users. Using one of several VoIP providers, the user can make an unlimited number of calls for a single price and also use a host of other applications like caller ID, call waiting and others that require additional fees for users of traditional telephony services.

The Talking House application, based on Zultys Enterprise Media Exchange IP PBX and a text-to-speech engine from NeoSpeech, Inc., enables the company to type scripts with details on specific listings. Patrick Ferriter, vice president of product marketing for Zultys Technologies, said that the speech application provides the information to the caller in English or Spanish, depending on the voice prompt from the caller.

"The customized prompts make it easy for users of the application," said Ferriter, adding that the IP connectivity enables centralization of the speech application, rather than the need for one in each of the real estate company's three locations.

Before adding Talking Home, Coldwell Banker Elite real estate agents were left to make their own voicemail messages, the quality of which varied greatly. Additionally, the information on the messages themselves was different from agent to agent. Now they type the information into a standard script, so listings are consistent. The application ensures that voice quality is consistent as well.

"This standardizes the whole process for us," Breen said.

According to Breen, another benefit from the Talking Home application is the ability to track agents. Previously, he didn't know who was in or out of the company's three offices. Now he can simply check his computer. The VoIP solution lets Breen know who's in the office and logged into the phone system at all times.

According to Breen, who employs 150 agents, an unexpected advantage that CB Elite has enjoyed from the VoIP system is the ability to recruit top-notch talent. When agents from competing real estate agencies heard about CB Elite's use of the latest technologies, they proactively contacted the agency for job opportunities.

VoIP, Speech-enabled IVR Both Growing
Frost & Sullivan's Popova explained that the increasing popularity of VoIP could mean a spike in speech-enabled IVR applications in the next few years. Many companies replaced their IVR systems in the late 1990s. Most of those systems are now at the end of their useful lives.

Many companies replacing these systems will be moving to IVR systems that run over VoIP, and will be looking to add other enhancements, such as speech applications. According to Popova, "This is where speech applications will get the most traction. Most IVR vendors are voice-enabling their systems. In the meantime, they are also voice-enabling their IVRs."

IVR providers are in strong agreement.

"If you buy an IVR that isn't VoIP capable, you'll be fired," says G. Rampranad, director of product line management, 3Com.

Lawrence Byrd, director of communications applications, Avaya,  said of VoIP and speech technology: "The two technologies are very complimentary. There are no other two communications technologies that have this kind of leverage.  It's a case of one plus one equals three. By using the two technologies together, you get three times as much benefit as you would out of a single system. The key difference is that you're moving from hardware to software to run the application. No additional hardware is needed."

Betsy Wood, Nortel marketing manager, added that by using VoIP, an administrator in a call center can more easily manage and monitor all of the telecom and computer resources in the network. The customers benefit from the speech application.

"Speech technology is focused on providing better service when customers call a company," Wood said. The caller simply states the problem and the speech application routes him to the proper agent. For the customer, the speech application is much faster and more user-friendly than a touchtone system.

Byrd said that the speech-enabled IVR systems will run the older touchtone IVR scripts as well.

"With VoIP you can add speech applications more cost efficiently," Frost & Sullivan's Popova added. Therefore, as companies replace IVRs, the lower total cost of adding a speech application makes any ROI proposition much more attractive.

Beyond the relative simplicity of installing speech and other applications at one central location, VoIP also solves the problem of wiring telecommunications systems in remote locations. There is no need to add telephone wiring in addition to computer wiring. Once installed, the systems are easier to move as well.

"No one likes touchtone; they like speech," Byrd explained. "Speech provides richer customer experiences. It does a better job of getting you to the right agent in a call center."

For example, Marin County, Calif. government employees must maintain close contact with the community, even when their jobs take them far away from the office. They stay in touch by calling a single number and speaking simple voice commands into the phone. Using Avaya speech access and technology that links IP PBX's with traditional PBX's, employees listen to and compose e-mails, check and schedule appointments and reach contacts from any wired or wireless phone.

The county's employees also use preference-based capabilities - such as "find me" functionality and instant alerts - to gain one-number accessibility while maintaining complete control over their communications needs. Residents trying to call a county supervisor can choose to "find" them, and Avaya's advanced mobility solution will attempt to contact the supervisor at a designated phone number (i.e., home phone, cell phone) based on user-defined rules. Employees can set their preferences so that the application will alert them for urgent voicemail or email messages - based on the priority level or the sender. For example, Marin County Sheriff's Department supervisors with critical information or evidence can quickly reach deputies who are investigating a crime.

Frost & Sullivan's Popova cautioned that the growth in the deployment of speech-enabled IVRs will be gradual. Companies, especially telecom firms, are trying to squeeze as much useful life out of their legacy IVRs as possible.

Others have installed speech-enabled IVRs that can work within a VoIP environment, but have yet to join the two technologies.
 
Eric Harshman, voice operations manager for Palo Alto, Calif.-based HP, said the company added a speech-enabled IVR to its busiest call center a little less than two years ago. Though the Nortel IVR can work with the VoIP system that many of the company's salespeople use, the two have yet to be integrated. He expects to integrate the two as the Session Initiation Protocol (SIP) standards for IP devices become more evolved. Other companies operate IVRs over VoIP systems using the existing SIP standards.

Even so, the company is seeing some of the benefits of speech technology already. The software enabled quicker call transfers, improved the call center's efficiency and improved customer satisfaction.

"Before we installed the speech capability, we didn't have a menu tree, we had a menu forest," Harshman said. The speech capability, which uses the Nuance speech engine, has cut customer-waiting time from 80 seconds to 30 seconds while also making it easier to add or delete items, which happens on a weekly basis.

When the SIP standards evolve further, he envisions being able to link the speech-enabled IVR capabilities with Internet chat, video and voice portals.

Recording Boost
Mookie Tenembaum, founder of United Virtualities, New York, a company that provides HotRecorder, a software to enable voice recordings via IP-enabled devices, said that one of the newer applications is voice recording via VoIP.

Recordings of traditional voice calls require storage and review of several hours of tape to locate recordings, Tenembaum said. But recordings on VoIP networks enable the user to employ software searches of audio files using key words, phrases or other parameters, much like one would search for a word or phrase in a Microsoft Word document on a computer.

This is important for companies that need an accurate recording of a conversation for legal purposes or other reasons. "Sometimes companies need to prove that certain conversations happened in certain ways," Tenembaum said.

Tenembaum explained that the idea for the company's HotRecorder software came from a patent attorney who was using a Skype VoIP-enabled phone. The attorney couldn't remember everything in all conversations and traditional recording methods were too cumbersome.

According to Tenembaum, 175,000 users downloaded the application between February and May. Skype users were the primary downloaders.

Tenembaum said that IP phone conversation recording isn't the only use for the software. He expects demand for the application to grow as podcasting becomes more popular. Podcasters publish their podcasts via audio files posted on the Internet.

Podcast subscribers use the HotRecorder application to search audio files for specific programs or topics. The podcasts are currently free or have small subscription fees. The next frontier, according to Tenembaum, is to combine paid advertising with podcasts.

Popova said that beyond the above deployments, speech technology capabilities and VoIP delivery are also coming together in such applications as unified messaging.

For example, Condell Medical Center, an acute care health facility in Libertyville, Ill. uses Avaya wireless IP phones in conjunction with the nurse call system to improve responsiveness to patients. When a patient pushes the call assist button, a text message is automatically displayed on the attending nurse's wireless handset. Nurses immediately know that assistance is needed, no matter where they are in the hospital. Similarly, Condell staffers can send text messages to members of its environmental services team to let them know when a room needs to be cleaned or serviced.

Popova expects further integration of the two technologies as they continue to evolve. "There isn't a single strong inflection point for massive growth in the market, so there won't be a 'hockey stick,' jump," she said. "We expect both these things to grow gradually; they won't take off over night."


Philip Britt is president of S&P Enterprises, Inc., an editorial services firm.  In that capacity, he has covered telecommunications, banking technology and related subjects for more than 10 years for several national publications. He has also been on show daily staff for SUPERCOMM for the last five years.

Page1 of 1