Omilia automates customer service in the dialog mode — in the call center, messengers, social networks, online chats on the website or in the application and allows you to completely replace the operator in the vast majority of cases of contacting the contact center. Modern technology of speech recognition and interpretation of meaning simulates a natural non-linear dialogue, taking into account the peculiarities of speech behavior of a living person. Omilia's artificial intelligence extracts meaning from customer requests and makes decisions automatically.
The platform allows you not only to reduce the cost of maintaining a contact center, but also to improve the quality of service: the absence of a multi-level voice menu, the elimination of erroneous transfers to the skill group and the reduction of waiting time — your customers can receive the necessary information or service almost instantly!
DiaManT is the core of the platform and a single point of integration with back office systems. DiaManT processes information received through all channels of interaction with the client — text and voice. It is in this part of the platform that the mechanisms of natural speech recognition — deepNLU (Natural Language Understanding), based on artificial intelligence, are concentrated. deepNLU analyzes the vocabulary, grammar, and semantics of the text, and then performs a conclusion from a pre-built database of facts and rules in accordance with the laws of formal logic. In addition, the system takes into account the intonation of the speaker, which also affects the robot's reaction: an incoming call from an agitated or dissatisfied customer will be transferred to the operator as quickly as possible.
DiaManT remembers the history and context of the client's requests, which ensures continuous communication and the possibility of a seamless transition from one channel of interaction with the client to another.
Deep ASR (Automatic Speech Recognition) – a component for automatic speech recognition. This part of the system converts the voice message into text and transmits it to DiaManT.
Deep ASR uses neural networks and built-in noise reduction mechanisms, which ensures high accuracy of speech recognition. Deep ASR supports a complex language model and is able to work with multiple languages at the same time, as well as with their combinations, dialects and accents.
Deep VB is a component for voice biometrics. In the process of communicating with the client, the system creates and stores unique "voice prints", which are used for passive verification of the client during subsequent calls. Depending on the security requirements for a particular client request, Omilia uses voice biometrics as an additional or single authentication factor, and is able to dynamically return to basic security issues (i.e. PIN, date of birth, etc.). Thus, voice biometrics allows you to simplify and speed up customer service.
Voice prints are stored in a closed format in the system's databases and cannot be decompiled and used outside the system. All personal and financial data is stored in accordance with the requirements of the international payment card data security standard PCI DSS for a limited period of time in an encrypted format in the on-premises enterprise data center or on the Omilia cloud.
OmIVR (Interactive Voice Recognition) - a component for receiving and distributing the flow of incoming calls. OmIVR addresses Omilia's speech recognition and synthesis systems, providing high-speed and accurate call routing.
OmIVR runs on both physical and virtual hardware and can be scaled to meet the needs of the client organization.