A lot of companies that have large list of clients organize clients support and servicing by phone. These services have a lot of employees that are constantly answering lots of questions and as a result maintenance and support costs for these services are expensive. For decreasing expenses for support services can be developed special automatic phone call processing systems which will process call without operator. Call will be redirected to operator only if the automatic system wasn't able to help. This way can reduce expenses for organization of any clients support systems.

Main part of this system is the service which provides speech assistant features to callers for persons that use a regular or software phone for connecting to corporation telephony hardware. The speech application is able to control the call state. It is able to interrupt the call and to get notifications when the call is being answered. The speech application prompts the caller and recognizes the input to perform necessary actions. Application provides menu dialogs, login features and finally service. These systems are able to deal with a lot of problems: customer account information access, product locator, troubleshooting guides, ordering systems, employee or customer PIN/password reset.

These systems have voice recognition and voice generation core. Modern text-to-speech engines allow generating natural speech dealing with different languages. Voice recognition systems also allow recognizing voice of a calling person “on-the-fly”. That means there is no need to make a pause between words and to try to pronounce words clearly. The system is able to cope with the usual human speech. Basically these systems based on Microsoft Speech Server and .NET technology.