Sabin's bilingual robot reduced the load on Sberbank's contact center in Kazakhstan to 26%

2021: Creating the Sabina Robot

The CST Group (part of the Sber ecosystem), on April 14, 2021, provided Sberbank Kazakhstan with its platform for creating intelligent robots based on artificial intelligence - Customer Engagement Platform.

Sberbank Kazakhstan created robot SABI (System Artificial Bank Intelligence), which is designed to accelerate work Contact Center and improve the quality of service. Voice assistant was launched in the summer of 2020, and until April 2021 allowed to reduce the load on the contact center to 26%. The robot advises customers telephone in both text channels on issues,, and crediting money transfers payment cards deposits. One of the features of Sabina is its bilinguality: the robot can communicate in two languages ​ ​ - Kazakh and Russian. The robot has been launched relatively recently, its skills are being improved. The peculiarity of the project is that three CST products based on artificial intelligence are used here at once, and:, neural networks and machine learning speech analytics voice text,. robots biometrics

The key idea of ​ ​ introducing a robot is the creation of a smart contact center controlled by artificial intelligence, the transformation of traditional customer service mechanisms into a multifunctional system that allows you to implement an individual approach to each client, "hearing his voice."

The development of Sabina includes three evolutionary stages:

"Consultant Navigator" - provides the client with general information on banking products and services, and, if necessary, redirects the client to the desired group of operators. This for April 2021 has already been implemented.

"Service Manager" - performs the functions of a consultant, presents service services by bank systems, such as voice biometrics, card blocking, issuing credit applications, managing payments, transfers, etc. This function is in the active phase of implementation as of April 2021.

And the third, the final stage of Sabina's evolution, will be realized in the future - "Personal Manager" - when the client's experience, behavioral model and client preferences become the main paradigm and motive for interaction for the robot, taking the bank's service and communication with customers to a different level.

The voice robot made it possible to reduce the call time to the bank's contact center, expand the channels of one-time customer service. For April, 2021 Sabine serves natural and legal entities in voice and not voice channels of service, including mobile prilozheniyasberbank online, Sberbusiness, advises clients not only on banking products, but also renders services of service character, including services for blocking of the card and exchange rates. The share of subscribers using the services of a voice robot has increased significantly. This was achieved, including due to the quality of synthesized speech of the CST group: the robot's voice is barely distinguishable from the voice of a living operator. CST group technologies allow the robot to quickly explore the digital landscape and instantly provide the user with a relevant response. The key quality of interaction with the CST group is the possibility of integrating an entire stack of AI solutions from one vendor, which has serious experience in enterprise-level projects: this allows you to be confident in the quality of products and their integration with each other. said Ernar Nakisbekov, Deputy Chairman of the Management Board of Sberbank Kazakhstan.

{{quote 'author = noted Dmitry Dyrmovsky, CEO of the MDG group of companies.|An omnichannel platform has been created that provides three-way interaction between the client, robot and operator. Sabina's difference is the prospect of integration with client voice analytics, lidogeneration and auto-calling systems: this will allow the robot to form a client profile taking into account existing communication experience, to carry out personal communication, knowing the client's preferences. The received answers can be used to self-teach the robot, improving each future dialogue. With Customer Engagement Platform, we help large contact center managers and IT directors reduce contact center operating costs, improve customer service, and increase sales.

Our algorithms allow you to recognize and synthesize speech, analyze 100% of client-operator dialogs, identify key causes of calls and answer 90% of typical questions in automatic mode, without the need to use human resources. The project implemented together with colleagues from Sberbank Kazakhstan demonstrates this.}}