Abstract: An approach for optimizing a confidence score threshold that is used to recognize a target word(s) in an audio source. A variety of potential instances of the target word can be detected and classified using an initial confidence score threshold value. Each potential instance of the target word is audibly reviewed and validated by a user. After a determination of the correctness of each potential instance's classification, a different confidence score threshold value can be used to produce an updated set of classification results without requiring the user to revalidate the results. By using a variety of confidence score threshold values to produce various sets of classification results, an optimized confidence threshold setting can be determined for the identified target word based on minimizing errors in the various results. This value can then be applied for future analysis of the target word in an audio source.

Abstract: A real-time speech analytics system (“RSTA system”) detects speech during a call involving a contact center agent and a remote party. Upon detecting the speech, an event message is generated by the RTSA system and transmitted to a module, which is configured to generate a topic meta-data record stored in a topic meta-data file. A real-time topic dashboard indication may be presented on a graphical user interface to a user based on analysis of a plurality of topic meta-data records. Selection of the topic dashboard indicator may result in presentation to the user of further text-based information about detection of the speech event for one or more calls. Selection of the further information for a particular call may result in a corresponding portion of recorded audio of that call associated with the speech event to be retrieved using the corresponding topic meta-data record and played to the user.

Abstract: A real-time speech analytics system (“RSTA system”) detects speech during a call involving a contact center agent and a remote party. Upon detecting the speech, an event message is generated by the RTSA system and transmitted to a module, which is configured to generate a topic meta-data record stored in a topic meta-data file. A real-time topic dashboard indication may be presented on a graphical user interface to a user based on analysis of a plurality of topic meta-data records. Selection of the topic dashboard indicator may result in presentation to the user of further text-based information about detection of the speech event for one or more calls. Selection of the further information for a particular call may result in a corresponding portion of recorded audio of that call associated with the speech event to be retrieved using the corresponding topic meta-data record and played to the user.

Abstract: An accurate determination of the disposition of a call origination is made in a contact center by mapping Session Initiated Protocol cause codes to reflect whether a call attempt was made to a remote party. Certain cause codes reflect various conditions where a call may originate from a contact center, but for various reasons, was not offered to the remote party, or more specifically, to an interface associated with the telephone number of the remote party. Reflecting these calls as incomplete calls, as opposed to call attempts, allows the contact center to more accurately reflect the number of call attempts. Accurately tracking the number of call attempts is important to comply with various state and federal regulations regulating the number of call attempts during a given time period. Further, incomplete calls may be symptomatic of a facility failure requiring attention, causing an administrator to be alerted.