A sequence of call volume measurements is accessed, where each of the call volume measurements is associated with respective metadata. The respective metadata may provide information regarding a time period during which a call volume measurement was made. A window of the sequence of call volume measurements with the respective metadata is input to a machine learning model to obtain a prediction of a call volume. The machine learning model includes embedding functions that are applied to the respective metadata for the call volume measurements in the window.

A method for filtering a plurality of agent-customer interactions to determine whether one or more of a plurality of agent-customer interactions should be stored in a library of Artificial Intelligence (AI) files related to an interactive voice response system (IVR) is provided. The method may include receiving an identification of a plurality of IVR flashpoints, monitoring and/or reviewing the plurality of agent-customer interactions, and determining whether one of the plurality of agent-customer interactions meets a threshold number of the IVR flashpoints. For each of the plurality of agent-customer interactions that meets a threshold number of the IVR flashpoints, the method may further direct the IVR to convert the interaction into an IVR workflow and store the IVR workflow in the library of AI related to IVR.

A sequence of call volume measurements is accessed, where each of the call volume measurements is associated with respective metadata. The respective metadata may provide information regarding a time period during which a call volume measurement was made. A window of the sequence of call volume measurements with the respective metadata is input to a machine learning model to obtain a prediction of a call volume. The machine learning model includes embedding functions that are applied to the respective metadata for the call volume measurements in the window.

An apparatus monitoring calls of a system includes a processor and memory causing the apparatus to perform operations including capturing recordings of calls in real-time from recorded devices. The processor may further cause the apparatus to compare audio call data of the calls detected by a switch with audio content of recorded calls by recorder devices to determine whether the audio call data matches the audio content of recorded calls. The processor may further cause the apparatus to monitor conditions of recorder devices during calls to determine whether a component(s) of recorder devices or of communication devices exceeds a threshold. The processor may further cause the apparatus to monitor memory in which recorded calls are transferred for archiving to determine whether there is enough storage in the memory. The processor may further cause the apparatus to determine metrics impacting quality of recorded calls and perform analytics on the recorded calls.

A machine learning model (e.g., including a deep learning neural network) with learned embeddings is applied to time series data with associated metadata to obtain predictions of the time series value. For example, a call volume in a period of time may be predicted based on call volume data for a sequence of time bins in a window of preceding time. Time bins may be associated with respective metadata, such as day of week, hour of day, day of month, holiday, part of business cycle, weather, and/or tide. These pieces of metadata may be mapped to embedding vectors using trained embedding functions. The resulting embedding vectors may be input to a neural network along with the corresponding time series data (e.g., call volumes) to make a prediction for future time bin. For example, the prediction may be used to provision servers in a network infrastructure.

An apparatus monitoring calls of a system includes a processor and memory causing the apparatus to perform operations including capturing recordings of calls in real-time from recorded devices. The processor may further cause the apparatus to compare audio call data of the calls detected by a switch with audio content of recorded calls by recorder devices to determine whether the audio call data matches the audio content of recorded calls. The processor may further cause the apparatus to monitor conditions of recorder devices during calls to determine whether a component(s) of recorder devices or of communication devices exceeds a threshold. The processor may further cause the apparatus to monitor memory in which recorded calls are transferred for archiving to determine whether there is enough storage in the memory. The processor may further cause the apparatus to determine metrics impacting quality of recorded calls and perform analytics on the recorded calls.

A method for filtering a plurality of agent-customer interactions to determine whether one or more of a plurality of agent-customer interactions should be stored in a library of Artificial Intelligence (AI) files related to an interactive voice response system (IVR) is provided. The method may include receiving an identification of a plurality of IVR flashpoints, monitoring and/or reviewing the plurality of agent-customer interactions, and determining whether one of the plurality of agent-customer interactions meets a threshold number of the IVR flashpoints. For each of the plurality of agent-customer interactions that meets a threshold number of the IVR flashpoints, the method may further direct the IVR to convert the interaction into an IVR workflow and store the IVR workflow in the library of AI related to IVR.

A data aggregation service is configured to minimize the number of service calls made to network services. The data aggregation service might be configured to cache objects returned in response to service calls to network services in a shared data object. A hash of the input values in the service calls to the network services may be utilized to determine whether an object referenced by the cache can be utilized instead of making another service call. The data aggregation service might also be configured to utilize change tracking to determine when calls are to be made to dependent services. The data aggregation service might also be configured to utilize data-specific time to live (TTL) values, to allow network services to specify the data they are interested in at a granular level, and/or to perform automated optimization of TTL values. Other optimizations might also be implemented by the data aggregation service.