Fighting Against Call Center Fraud with Voice Biometrics

Voice Biometrics for Call Centers

The number of attempts to defraud organizations is increasing each day. Today’s fraudsters are skilled in social engineering—the use of deception to obtain confidential information—and use sophisticated methods. Thus, they can easily gather different kinds of data to use in attacks like identity theft, account takeovers, and fraudulent transactions.

When compared with better-controlled digital channels, call centers are at greater risk. Their dependence on a human workforce increases their vulnerability to fraud. Employee training generally falls short when it comes to sophisticated social engineering attacks, and it is not hard to answer security questions by using compromised customer data. This encourages fraudsters to target call centers.

What Can Call Centers Do?

Traditional knowledge-based authentication methods like challenge questions are no longer enough to prevent fraud. Call centers need to find efficient methods to protect themselves from increasingly sophisticated attacks. This requires a holistic security approach where security and risk management are taken seriously, which means investing in fraud prevention technologies that take advantage of the power of biometrics.

Why Voice Biometrics?

For many organizations, voice biometrics is an effective security measure, and the use of this technology is constantly increasing. According to a recent market study, the global voice biometrics solutions market is expected to reach a value of US$13 billion by 2026, with a compound annual growth rate of 17.3%.

This means that more and more companies are applying voice biometrics to their security needs, and call centers are no exception. Call centers prefer this technology due to following reasons:

It is the only biometric measure for call centers

Voice biometrics is the only biometric measure that can be used over the phone, which makes it literally the only biometric option for call centers, but organizations use it because of its many advantages, not because it is their only option.

Biometric technologies differ from common security measures by being based on unique individual characteristics like the retina or iris. Unlike knowledge-based security methods, a person’s voice can’t be compromised. By analyzing a user’s voice for its hundreds of unique characteristics, voice biometrics offers a convenient and highly secure authentication method. Call centers that aim to implement the latest security technology will choose biometrics, which is really the only option for call centers.

Other methods might fall short in protection

Traditional security measures include a “what you know” factor. PINs, passwords, and mother’s maiden names are all examples of knowledge-based verification questions. Even the most advanced security questions are often based on public information that can easily be obtained by a fraudster. Therefore, knowledge-based authentication doesn’t offer an effective solution against fraud.

On the other hand, biometric verification boasts a “what you are” factor because is based on an individual’s unique characteristics, including voice. A voice can’t be stolen, hacked, or forgotten, which is why voice biometric software offers a more secure authentication method than the alternatives.

Voice biometrics can track fraudsters’ voices as they call across multiple accounts and time periods. Although fraudsters can trick agents, they can’t fool voice biometrics. Thus, the technology reduces call centers’ vulnerability to social engineering attempts.

It can be combined with other security methods

Today’s security-concerned organizations use multi-factor authentication to overcome potential threats. They layer several authentication methods together to deliver the strongest possible security solution. Voice biometrics enables multi-factor security by integrating with a variety of security methods.

If an organization already has a preferred identity verification method, voice biometrics can enhance this method by providing an added element of security. This adds a “who you are” layer to the traditional “what you know” layer of passwords and PINs. For example, single-factor authentication, such as simply entering a password, has been proven to be a weak solution. On the other hand, a password coupled with voice biometrics can guarantee secure authentication.

It doesn’t sacrifice the user experience

Conventional security measures like PINs, passwords, and security questions can be easy to forget and time consuming for customers. Voice biometrics saves customers from the need to remember passwords and the burden of invasive knowledge-based questions.

The technology enables identity verification in a matter of seconds by analyzing users’ individual voiceprints. Seamless and consistent security not only removes frustration but also saves time both for customer and agent. Therefore, voice biometrics reduces stress, saves time, and greatly improves the customer experience.

It supports multiple channels

Thanks to the omnichannel approach, today’s customer interactions have moved beyond call centers. Customers can reach organizations from whatever channel they prefer, including IVR, online, social, and mobile.

Voice biometrics is a natural fit for cross-channel engagement. The technology offers efficient verification in a way that enhances the user experience and provides consistent multi-channel security.

It offers flexibility

Voice biometrics can be deployed on the premises or in a variety of cloud settings to meet organizations’ different security needs. Thanks to the highly customizable structure of the technology, any organization can set up a fast, easy, and powerful security solution within its existing platforms and technologies.

Voice biometrics also offers accent- and language-independent use; the technology verifies user identities strictly by voice, not by their language or accent. This provides flexibility for organizations by allowing them to implement this technology into any operation regardless of location.

It reduces business costs

Voice biometrics significantly reduces the number of security steps and the time involved in the verification process. Agents spend less time with each caller, shortening average call times. This not only saves call centers significantly on overhead costs but also results in happier customers and a more efficient team.

Voice biometrics also reduces the number of fraud attacks by detecting repeat calls from known fraudsters. This contributes to remarkable savings in avoided fraud incidents.

Use Cases: Fraud Protection with Voice Biometrics

Text-Dependent Voice Verification

Text-dependent voice verification offers an active verification method. This technology requires passphrases to enroll and authenticate users. First, users create a voiceprint during the enrollment process by repeating a specific passphrase. In their subsequent calls, they simply repeat this passphrase to verify their identities. The voiceprint of a caller speaking the passphrase is compared with prior voiceprints for authentication.

The technology offers effective security against fraud. It is impossible to impersonate or fool the technology because the user’s unique voice details must be matched with the voiceprint on file. Features like playback manipulation detection, voice change detection, synthetic voice detection, and blacklist identification enhance the power of this technology to combat various security threats.

For example, with playback manipulation detection, the technology recognizes that a captured voice is being played. Voice change detection recognizes when the end user’s voice changes during the enrollment process, and synthetic voice detection identifies cases in which fraudsters modify a recorded voice sample. The technology also checks for known fraudsters by employing biometric blacklist identification during all operations.

Visit the Sestek Vocal Passphrase page to learn more about our text-dependent voice verification technology.

Text-Independent Voice Verification

Text-independent voice verification offers a passive verification method. With this technology, callers don’t need to take a specific action to pass through an authentication process. Unlike active authentication, passive authentication doesn’t require formal enrollment or the repetition of a specific passphrase during authentication.

To use the system, callers must first enroll by calling the call center and speaking. After a customer speaks for a sufficient number of seconds, the system creates a voiceprint that will be used to authenticate them on their next call. Simply by listening to customers talk with agents, the system scores the recording’s relevance by comparing it with the previously enrolled recorded voiceprints. If the score is high enough to accept the user, the security check is passed.

This technology provides the benefit of enhancing fraud protection without troubling users. With such an effortless authentication experience, the technology not only boosts customer satisfaction but also ensures biometric security. Features like playback manipulation detection, synthetic voice detection, and blacklist identification ensure that the technology will not be fooled by fraudsters.

Visit the Sestek Verification On-The-Go page to learn about our text-independent voice verification technology.

Fraud Detection

Biometric fraud detection solutions are used to identify and stop highly specialized fraudulent individuals. Blacklist applications are among the most common fraud detection methods. Based on voice biometrics, these applications detect fraudsters through biometric verification and trigger alarms for anti-fraud teams.

Biometric blacklist application seamlessly identifies users with high fraud potential, then diverts them into a separate queue for call security teams. The application automatically crosschecks every call with the blacklist database after a voiceprint is recorded. It marks those calls that are scored higher than the threshold score and identifies listed fraudsters by their voices, comparing the captured voice with voiceprints stored in the blacklist voiceprint database.

Visit the Sestek Blacklist Identification page to learn more about our biometric blacklist application technology.


Call center fraud is a widespread problem. Depending on human agents increases call centers’ vulnerability to fraud, and fraudsters try to take advantage of the human factor by using different techniques to break through the relatively simply barriers that call centers typically have in place.

Call centers no longer rely on traditional knowledge-based authentication. They know that overcoming constant security threats requires a more effective solution. Therefore, more and more organizations are turning to voice biometrics for effective fraud protection, an enhanced customer experience, and higher call center efficiency.

Author: Hilal Bakanay, Senior Marketing Specialist, Sestek