The Journey of Speech Recognition IVR

In the last 15 years, technology has advanced rapidly, delivering everyday electronics we take for granted including smartphones, consumer Satellite-Navigation systems and Apple’s iPod.

Although we are so accustomed to the rapid development of consumer technology, speech technology has been fairly slow to overcome hurdles.

A Brief History of Speech Recognition

With the first electronic speech synthesiser dating back to 1936, speech recognition has been developing in the background for a very long time.

The first telephone that could dial dual tone multi-frequency (DTMF) audible tones was unveiled in 1962 by Bell System. The plan for Interactive Voice Response (IVR) was born. With the development of IVR technology, call centres are now able to process voice recognition as well as DTMF tones followed by an intelligent queueing and routing decision.

In 1999, Brian Bischoff, general manager for AT&T Solutions Business Development said in the near term, speech recognition provides a bigger gain than even the Web. "That's how customers contact and interact with businesses." (Information Week).

As this basic timeline reveals, speech recognition has taken quite a while to get off the ground, originally costing thousands of pounds and suffering from factors such as lack of accuracy, sensitivity to noise and generally being very difficult to operate. Moving onto figures from 2014, according to an IBISWorld Speech & Voice Recognition Software Developers market research, the voice recognition software industry was worth around £14billion a year in revenue in 2014, growing annually at 10.7%.

So what has changed the game for speech recognition?

A fundamental change in speech recognition has been consistently improving its accuracy. Now able to achieve accuracy percentages in the high 90's, the experience of using speech recognition is much more fluid for a user.

The integration of speech recognition has also become commercially widespread, featuring in Windows operating systems, mobile devices (such as Siri, Google Voice) and used in time-sensitive environments such as medical documentation, military and aerospace. With mainstream devices such as iPhones bringing speech recognition to the spotlight through Siri, we've seen the technology spread across everyday devices such as televisions, wearable technology (Google Glass and Google's Smart Watch) and even cars.

In the contact centre industry, speech recognition has become a foundation of self-service interactive voice response (IVR) user interfaces. With increasing call volumes and therefore rising employment costs, contact centres have found speech recognition IVR applications to be cost-effective and the friendliest, speediest self-service alternative to speaking with a contact centre agent.

How does it work in the Call Centre environment?

By 2020, the customer will manage 85% of the relationship with an enterprise without interacting with a human. (Source: Gartner).

Current IVRs give callers access to general enquiries services such as automated film times, airline schedules, place orders or even just simply use a directory to contact a department or individual.

The level of complexity is illustrated in the following examples of user interaction with a speech-recognition IVR:

Touch-tone replacement
System Prompt: "For your current account information, press or say one."

Directed dialogue

System Prompt: "Would you like current account information or rate information?"
Caller Response: "Current", or "Current account", or "rates."

Natural Language
System Prompt: "What transaction would you like to perform?" Caller response: "Transfer £500 from current account to savings"

The benefits?

Other than the obvious operational costs of providing an IVR to handle calls, speech recognition has always been viewed by customer services as more appealing as a more 'friendly' interface for an automated interaction. In addition, companies benefit from:

1) Reduced Call Queues - agents are given more time to handle more complex enquiries with a real conversation, rather than rushing through calls to get through the volume.
2) Shorter calls – owing to the higher use of interaction with an IVR.
3) Hands-free - Being able to talk to a phone system is more personal and more flexible.
4) More efficient - An efficient speech recognition system allows for a much faster customer resolution.

The future of Speech Recognition

Here at Eckoh, we are able to boast a 99% proven success rate for our speech recognition IVR system; EckohASSIST. This system uses Natural Language speech recognition as described above, which eliminates the need for complex menu systems and takes the caller straight to the destination. With that level of accuracy, we know speech recognition will be soon be used as a major customer service enabler for businesses around the world. Businesses are now realising just how far speech technology has come in just the last five years, and now trust it to their customers to aid their journeys.

As consumers integrate speech recognition technology into their everyday lives, they are becoming familiar with giving instructions verbally to a machine. In a short time, it will be generally expected that businesses integrate this as a form of customer communication.

Speech recognition beyond 2010:

2010- Microsoft Kinect
2012- Samsung TV
2012- Google Glass
2014- Android Wear
2014- Amazon Echo
2015- Apple Watch

Future speech recognition predictors:

- Self-Driving Car Interface
- Virtual Reality Voice Recognition Entertainment
- Flight Controls
- Voice Biometrics
Loading Conversation

Posted by eckoh at 11:42 AM on Sep 27, 2016


Recent Posts

Does it feel like it's panto season in your contact centre every day? If so,…


We've all watched on in horror ... as an out-of-control child creates havoc at a…


Are customers mysteriously falling out of love with your business — despite your…