Progress Made Toward Speech-Based Predictive Analytics

Typical business analytics solutions have traditionally relied on demographic and historical data to draw their conclusions, but a new type of solution using voice analysis could hold the key to helping companies better predict customer behavior.

VoiceSense, a provider of predictive analytics for enterprises, recently launched SEAL (Speech Enterprise Analytics Leverage), a speech-based solution that brings an innovative new approach to Big Data predictive analytics by focusing on the behavior tendencies of customers rather than their demographic and historical information. Through the analysis of prosodic (non-content) speech parameters such as intonation, pace, stress levels, and more, SEAL can accurately predict future consumer behavior.

“The use cases for this technology are almost limitless,” says Yoav Degani, VoiceSense’s CEO. “There is a huge market for data analytics—$200 billion per year.”

Degani says his company’s product offers a “unique approach.”

“We are bringing speech into data analytics,” he states. “This gives [users] a different angle.”

According to Degani, the new solution shortens and improves decision-making processes. Additionally, it can be used for a wide variety of voice-based customer interaction scenarios to tackle a range of goals: improve risk assessment, expand sales, and increase customer retention. The solution also could perform health and wellness monitoring and assist human resources departments with personal profiles of job candidates and staff.

SEAL can be used to automatically analyze recorded voice files or live audio streams and is completely independent of language and content. Predictive models and signal processing techniques are used to assess more than 200 prosodic speech parameters.

The solution uses the analysis to build personal profiling models while evaluating an individual’s characteristics, including levels of risk affinity or aversion, tendencies for impulsive behavior and rule abidance, personal integrity, conscientiousness, sociability, and well-being, to develop prediction scores for specific consumer behaviors that are then automatically integrated into decision-making processes and CRM data.

“Predictive speech analytics can help organizations monitor and anticipate customer behavior, enabling them to identify opportunities and provide personalized service,” explained Raul Castañon-Martinez, a senior analyst at 451 Research, in a statement. “The range of use cases that can be addressed make it particularly relevant for high-touch customer interactions, which can have a direct impact on a company’s bottom line.”

VoiceSense recently completed pilot implementations of the SEAL solution with a number of banks and insurance companies in the United States, Israel—where VoiceSense is based—and parts of Asia and Australia. During these pilots, SEAL delivered accurate loan default predictions and insurance claim probabilities by analyzing recorded phone calls. SEAL’s accurate predictions were confirmed by the actual outcomes experienced by these enterprises.

“They are all offering loans, but they are concerned about the risk,” Degani says. “If we can help them with their risk, that is huge. If they can approve an online loan without waiting for paperwork, they can increase their business about 10 percent.”

Degani also sees the solution being used to help confirm insurance claims, as it did for an Australian insurer during the pilot.

“After many years of research, we are proud to be launching our game-changing solution for uniquely and accurately predicting customer behaviors for which the markets and use case scenarios are almost limitless,” Degani adds. “We are tremendously pleased with the initial market feedback, and we have strong expectations for our SEAL solution, as the value for any enterprise is enormous.”