Call Centers

Advanced Voice Response Tech Leading to Improved CX

speaking into smartphone

Multi-industry adoption of secure voice interactions in call centers will raise the bar on the reliability of voice authentication, fraud detection, and call intelligence technology.

Contact centers have been challenged the past year, most notably from labor shortages that have forced the industry into changing how it operates, how it staffs, and how it improves customer satisfaction and experience.

In response, many in the industry are turning to different methods that help automate their operations. Two key tools are interactive voice response (IVR) and intelligent voice assistant (IVA), which use artificial intelligence to enable self-service.

To that end, Pindrop in January announced its partnership with Five9, an intelligent cloud contact center provider, to make Pindrop’s security solutions more accessible across the globe. The cooperative venture will help secure customer voice interactions across industries.

Pindrop’s voice authentication and fraud detection technologies will be available later this year on Five9’s CX Marketplace, according to Walt Rossi, vice president of business development at Five9.

A key focus for Pindrop is to make voice authentication technology more beneficial to businesses across a variety of industries, added Gahn Lane, vice president for global partners and alliances at Pindrop.

The partnership comes as the call center industry is grappling with two major challenges. First, fraud and cybercrime attempts are increasing. Second, call centers are dealing with increased user traffic, which threatens to impact the quality of customer experience across the industry.

“Now more than ever, contact centers are hyper-focused on improving both user security and customer experience,” said Rossi.

What To Expect

Pindrop solutions help protect some of the biggest banks, insurers, and retailers in the world using patented technology that extracts intelligence from the calls and voices encountered.

The company hopes to establish the standard for identity, security, and intelligence for voice interactions.

Pindrop solutions also help detect fraudsters and authenticate genuine customers. This reduces fraud and operational costs while improving customer experience and protecting brand reputation.

Partnering with Five9 will expand Pindrop’s customer reach in key sectors such as retail, healthcare, and higher education.

The technology looks at everything from the acoustic characteristics of the carrier, the characteristics of the call itself, the handset, and even the landline or type of mobile phone connecting with the call center, explained Collin Davis, CTO of Pindrop.

“The plan for the technology was initially how we can help financial institutions, especially during Covid. Banks and financial institutions, security, and insurance companies saw an increase in traffic to their contact centers. Pindrop technology also now incorporates voice biometrics to be able to detect fraud,” Davis told CRM Buyer.

How It Works

Caller authentication technology routes out fraudulent callers. That, in turn, enables a better customer experience with shorter hold times on the phone. It also enables company representatives to not need to waste time sorting through security screening questions, Davis said.

The detection technology can identify the type of hardware on the incoming call and weed out calls, for instance, coming from Nigeria pretending to be over Skype from a different location. The technology also detects characteristics that do not match those of a valid caller.

Every business that has a complex number sends it to interactive voice responses for support. That is typically a contact center platform.

Businesses use these to route calls to their agents, and these go over normal telephone lines. Pindrop has voice biometric capabilities in the form of application programming interfaces or APIs that integrate with the call centers and Pindrop’s cloud-based platform.

One aspect of the technology involves carrier integration. We have partnerships with many phone carriers, so we get a copy of the call routed to us while the call contact center is hosted by the business, said Davis.

“It is highly reliable without any risk of cognitive interference with the contact center, and we are able to process the call as it is happening. We send information to our cloud infrastructure that allows the agent to be able to see the analysis of the call in real time,” he explained.

This approach lets Pindrop’s platform users modify the way that they handle and route calls based on this detection analysis if fraudsters call, he added. This helps businesses route fraud calls directly to fraud specialists as opposed to one of their agents.

Flexibility Expands Use Cases

Other scenarios, such as internet of things (IoT) connected devices, voice-enabled automobiles, and home alarm systems, enable businesses to identify user voice patterns. This opens up the possibilities for voice interaction.

“We started in the context we are continuing to support. The contact center is a natural, very obvious voice first-use case where if you are on the phone, you are using your voice, so it works really well,” said Davis.

However, with the increasing adoption of voice assistants and IoT devices, people do not talk to computers, so not all systems interact with the ability to understand who is doing this.

“It is a really powerful unlock for a lot of these different scenarios,” he noted.

Everything Pindrop is doing is based on machine learning and artificial intelligence that looks at various attributes of a caller’s voice, the type of your device, and the acoustic characteristics.

It detects tell-tale traits in the way a person speaks, the tonal effects from the throat and nose, and the way that the tones come out of your mouth.

“Characteristics and inflections that you have in your voice enable us to very accurately tell that this is your voice,” Davis said. “Even the age or the gender of the speaker [plays a role]. We are getting to the point where we can also tell things like if this individual speaker is agitated or excited.”

Jack M. Germain

Jack M. Germain has been an ECT News Network reporter since 2003. His main areas of focus are enterprise IT, Linux and open-source technologies. He is an esteemed reviewer of Linux distros and other open-source software. In addition, Jack extensively covers business technology and privacy issues, as well as developments in e-commerce and consumer electronics. Email Jack.

Leave a Comment

Please sign in to post or reply to a comment. New users create a free account.

More by Jack M. Germain
More in Call Centers

CRM Buyer Channels