The Adaptive Audio – VUI Cloud Blog

During production trials from March 8, 2012 through March 16, 2012 adaptive technology improved the success rate of automated calls at a client site by 1 percent.

The client, an unnamed IVR hosting center that supports thousands of ports and dozens of voice applications, segregated out select groups of adaptive and non-adaptive ports for exact performance measurements. The adaptive ports were using Interactive Digital’s Adaptive Audio (www.interactive-digital.com) software.

As Figure 1 below shows, the call center received a total of 37601 calls on the non-adaptive ports during this period while 37811 were received on the adaptive ones. Of those calls, respectively 10110 and 10545 were considered by the client to be successful. To meet the criteria for success, a caller had to receive information (bank balance, airline flight information, medical claim details etc.) and then either hang up or continue their call with an agent.

Speech is one of the few technologies that has the potential for enhancing telephone self-service.

The technology, if used properly, provides motivation for significant upgrading of existing self-service installations.

According to ASR News, the speech market will continue to grow at a healthy clip, as illustrated in Figure 1 below.

Figure 1 Telephone Self-Service Usage Projections
(Billions of Minutes)

In 2010, the total self-service installed base was 8.6M ports worldwide. Figure 2 shows the geographic breakdown for these port deployments.

Read more…

The hallmarks of a great self-service voice application are not a whole lot different from that of a great CSR. After all, they serve (or are supposed to serve) the same purpose; your customers need for service and information via telephone. In fact, from the caller’s perspective, engaging with an IVR application is similar to engaging with another human. Callers are, generally speaking, fair and reasonable individuals that are willing to work with a CSR or technology as long as that engagement seems to be working for them.

Figure 1 shows how an “engagement threshold” exists for the caller when they use an IVR application the first few times. Notice that, though it may be slightly different for Speech and DTMF, there is a significant threshold after 2 – 3 successful interactions by the caller. This critical period is when callers ask themselves whether this automated process will work for them so they don’t have to wait for a CSR, or whether they are wasting their time when they may end up talking to a CSR anyway.

Figure 1 - The Caller Engagement Threshold

Reaching this “buy in” point is critical to the success of any voice application. The likelihood that a returning caller will use the Read more…

In an earlier post and supporting white paper I talked about how caller adaptive technology helps the mobile experience and the handling of callers of different skill levels. Here, I want to present supporting data for this type of technology – data collected from production trials.

Based on production metrics gathered at various installations,the figure below illustrates the relationships between average call length, the number of script levels and the effectiveness of adaptive technology at optimizing the call process.

These data indicate improvements in the Average Handle Time (AHT), Average Handle Rate (AHR), IVR Utilization (IVR turns per call) and Caller Input Error Rates.

The benefits of adaptive technology to optimize self-service telephone calls vary based on the design, content and average call duration of Read more…

More than ever in the past, today’s telephony based voice applications must address large and very diverse calling populations.

This audience uses a wide variety of personal, mobile and landline based devices in various modes to access information over the phone.

Some facts and figures:

• At the end of 2011, there were 6 billion mobile subscriptions, according to the International Telecommunication Union.

• Mobile subscriptions outnumber fixed lines 5:1 (more so in developing nations)

• According to the US Census Bureau, the hearing loss population in the US has grown to over 31.5 million

• Linguists estimate that half of the world’s population is bilingual and there are about 5,000-6,000 different languages spoken in the world today.

• People of a particular culture or regional dialect may be comfortable with long pauses at the conclusion of a sentence, while people of another may find this demonstrates disinterest, lack of attentiveness and even disrespect.

Mobile phone use globally in 2011

Read more…

In an earlier blog entry, I mentioned some production tests we performed at a client site to measure optimal speaking rates (words per minute) for IVR callers.

Individual callers speed and accuracy determined whether (and by how much) the audio playback rate was increased or decreased at each IVR script Interaction Point (IP).

An IP might be as simple as answering yes/no, selecting from a menu, entering a 16 digit credit card number or anything in between.

A caller that behaved as an expert throughout the call would experience an audio playback speed adjustment profile that went from 100% to 108% and then to a final maximum rate of 112% of normal playback.

A caller that was not quite an expert, but better than average, experienced an adjustment profile of 100%, 105%, 108% and 110%.

Callers with less than average speed and accuracy experienced similar adjustment profiles in the opposite direction – 100%, 95%, 92% and so forth.

Read more…