A server can receive a plurality of records at a databases such that each record is associated with a phone call and includes at least one request generated based on a transcript of the phone call. The server can generate a training dataset based on the plurality of records. The server can further train a binary classification model using the training dataset. Next, the server can receive a live transcript of a phone call in progress. The server can generate at least one live request based on the live transcript using a natural language processing module of the server. The server can provide the at least one live request to the binary classification model as input to generate a prediction. Lastly, the server can transmit the prediction to an entity receiving the phone call in progress. The prediction can cause a transfer of the call to a chatbot.

The present disclosure relates generally to network diagnostics, and more specifically to techniques for determining the health of customer premises equipment (CPE) devices. In certain examples, a scoring system can determine a health score for a CPE device, the health score indicating the health of the CPE device. The health score is determined by receiving, analyzing, and integrating information from different sources, including sources in addition to CPE diagnostic data for the CPE device at the current time. Such sources can include CPE diagnostic data for the CPE device at other times, CPE diagnostic data for other CPE devices related to the CPE device (e.g., in the same household or surrounding area as the CPE device), technical specifications of the CPE device (e.g., make, model and firmware of the CPE device), or per-equalization parameters obtained by the CPE device.

A server can receive a plurality of records at a databases such that each record is associated with a phone call and includes at least one request generated based on a transcript of the phone call. The server can generate a training dataset based on the plurality of records. The server can further train a binary classification model using the training dataset. Next, the server can receive a live transcript of a phone call in progress. The server can generate at least one live request based on the live transcript using a natural language processing module of the server. The server can provide the at least one live request to the binary classification model as input to generate a prediction. Lastly, the server can transmit the prediction to an entity receiving the phone call in progress. The prediction can cause a transfer of the call to a chatbot.

According to one embodiment, a method, computer system, and computer program product for detecting and repairing gaps within a call is provided. The present invention may include requesting a personal discourse gap recovery model (PDGRM) for participants in a call, where the PDGRM may be a machine learning model that models a user's speech patterns based on word collocation, dictionary, and speaking rate. The present invention may further comprise detecting one or more gaps in the call based on network connectivity, inserting an audio delay into the call, constructing repaired dialogue to fill at least one of the one or more gaps in the call based on the PDGRM, and inserting the repaired dialogue into the call.

A server can receive a plurality of records at a databases such that each record is associated with a phone call and includes at least one request generated based on a transcript of the phone call. The server can generate a training dataset based on the plurality of records. The server can further train a binary classification model using the training dataset. Next, the server can receive a live transcript of a phone call in progress. The server can generate at least one live request based on the live transcript using a natural language processing module of the server. The server can provide the at least one live request to the binary classification model as input to generate a prediction. Lastly, the server can transmit the prediction to an entity receiving the phone call in progress. The prediction can cause a transfer of the call to a chatbot.

The present invention relates to a service system supporting a voice call using a digital assistant device, a method thereof, and a non-transitory computer readable medium having a computer program recorded thereon, and more particularly, to a service system which performs a call between a digital assistant device and a terminal corresponding to a call opponent through a voice command of a user and supports call conversion between the digital assistant device and a user terminal through the voice command of the user, a method thereof, and a non-transitory computer readable medium having a computer program recorded thereon.

The present disclosure provides a method of measuring the effectiveness of an intervention in a hybrid fiber-metal access network. The effectiveness measure is determined in accordance with the improvement in the attenuation and the maximum achievable data rate. The effectiveness measure is used to determine whether a further network intervention is required.

Disclosed in some examples are methods, systems, and machine-readable mediums which provide for a communication monitoring system configured to automatically monitor and process communication records according to one or more error detection profiles to detect one or more error conditions described in the error detection profiles. The system may then automatically determine a network segment that is experiencing the error condition and acts to correct or avoid the error condition. For example, the system may instruct a conference communication service to re-route a caller over a different network segment, instruct a user's device to use a different network segment, send a message to a reporting computing device of the offending network segment, or the like.

A server can receive a plurality of records at a databases such that each record is associated with a phone call and includes at least one request generated based on a transcript of the phone call. The server can generate a training dataset based on the plurality of records. The server can further train a binary classification model using the training dataset. Next, the server can receive a live transcript of a phone call in progress. The server can generate at least one live request based on the live transcript using a natural language processing module of the server. The server can provide the at least one live request to the binary classification model as input to generate a prediction. Lastly, the server can transmit the prediction to an entity receiving the phone call in progress. The prediction can cause a transfer of the call to a chatbot.

A method of determining the location of a disconnection on a digital subscriber line, in particular a VDSL line, where the line has a number of nodes or connection points along it. A statistical model is generated from a population of lines that maps the loop (line) lengths of each of those lines obtained by line test measurements (such as single ended line test tracesSELT traces) against the corresponding loop lengths obtained from inventory data. The model is then used to determine a predicted loop length by mapping a measured loop length (taken from line test measurements) onto an inventory loop length using the model. Knowledge of the node positions on the line is then used to give a probability a line disconnect occurring at a given node by mapping the predicted loop length onto the node positions.