Abstract: Systems and methods for generating encoded text representations of spoken utterances are disclosed. Audio data is received for a spoken utterance and analyzed to identify a nonverbal characteristic, such as a sentiment, a speaking rate, or a volume. An encoded text representation of the spoken utterance is generated, comprising a text transcription and a visual representation of the nonverbal characteristic. The visual representation comprises a geometric element, such as a graph or shape, or a variation in a text attribute, such as font, font size, or color. Analysis of the audio data and/or generation of the encoded text representation can be performed using machine learning.

Abstract: Systems and methods for improving generated images based on image accuracy and aesthetics are disclosed herein. For example, an image tuning system can receive an image generation prompt describing a first image. The system can generate a second image. The system can compute an accuracy metric and an aesthetic metric. The system can determine whether the accuracy metric and aesthetic metric are less than respective thresholds. The system can generate a seed artifact. The system can generate a third image approximating the first image based on the seed artifact. The system can transmit the third image for display at a user interface.

Abstract: This document describes techniques, apparatuses, and systems for a personalized set of images that are customized to a specific user using his/her likeness. The personalized set of images are presented as a personalized advertisement on a display that is determined to be visible to the user. The personalized set of images can be generated by a telecommunications network operator to which the user is subscribed. An image of the user is retrieved from a user profile associated with the user by the telecommunications network operator. The user's face can be isolated from the image and included within the personalized advertisement. As a result, the personalized advertisement can better appeal to the user, which can increase the effectiveness and palatability of the advertisement.

Abstract: Systems and methods for generating encoded text representations of spoken utterances are disclosed. Audio data is received for a spoken utterance and analyzed to identify a nonverbal characteristic, such as a sentiment, a speaking rate, or a volume. An encoded text representation of the spoken utterance is generated, comprising a text transcription and a visual representation of the nonverbal characteristic. The visual representation comprises a geometric element, such as a graph or shape, or a variation in a text attribute, such as font, font size, or color. Analysis of the audio data and/or generation of the encoded text representation can be performed using machine learning.

Abstract: Systems and methods for generating encoded text representations of spoken utterances are disclosed. Audio data is received for a spoken utterance and analyzed to identify a nonverbal characteristic, such as a sentiment, a speaking rate, or a volume. An encoded text representation of the spoken utterance is generated, comprising a text transcription and a visual representation of the nonverbal characteristic. The visual representation comprises a geometric element, such as a graph or shape, or a variation in a text attribute, such as font, font size, or color. Analysis of the audio data and/or generation of the encoded text representation can be performed using machine learning.

Abstract: Systems and methods for generating encoded text representations of spoken utterances are disclosed. Audio data is received for a spoken utterance and analyzed to identify a nonverbal characteristic, such as a sentiment, a speaking rate, or a volume. An encoded text representation of the spoken utterance is generated, comprising a text transcription and a visual representation of the nonverbal characteristic. The visual representation comprises a geometric element, such as a graph or shape, or a variation in a text attribute, such as font, font size, or color. Analysis of the audio data and/or generation of the encoded text representation can be performed using machine learning.

Abstract: Systems and methods for remotely resetting networked devices is disclosed. The system can comprise a primary network connection and a secondary network connection to provide communications redundancy. The system can also include physical connections between networked devices to enable devices to be reset regardless of the type of error. The method can include monitoring one or more parameters of a device to determine what type of error is occurring and what action to take. The system can enable a first computer (or other electronic device) to restart applications, adapters, and services on a second computer (or other electronic device). The system can also enable the first computer to reset or restart the second computer. The system can enable networked computers to monitor and restore operation to other computers on the same network without human intervention.

Abstract: Described herein are techniques, devices, and systems for using a machine learning model(s) and/or artificial intelligence algorithm(s) to optimize testing of components of a system operated by a wireless carrier. For example, data generated as a result of executing a first test of a suite of tests may be provided as input to a trained machine learning model(s) to classify one or more tests of the suite of tests as having a particular characteristic. A to-be-executed test may be classified as likely to pass or likely to fail when executed, for example. An already-executed test may be classified as reliable or unreliable, as another example. Based on the classification of the test(s), the suite of tests may be modified to optimize testing of the wireless carrier's system.

Abstract: Systems may comprise a connectivity manager application operating on a User Equipment (UE) for optimal and secure performance of local applications and improving user experience by optimizing application performance (e.g., by prioritizing file downloads). The systems may modify data transmission of other local applications on the UE in a multi-network environment. The connectivity manager application may detect one or more network characteristics of one or more networks providing service to the UE and determine a connectivity rule based at least in part on the one or more network characteristics and/or particular data types and indicating that data transmission for the one or more transmission paths is to be adjusted, modified, paused, or prohibited in response to particular network conditions, etc.

Abstract: Described herein are techniques, devices, and systems for using a machine learning model(s) and/or artificial intelligence algorithm(s) to optimize testing of components of a system operated by a wireless carrier. For example, data generated as a result of executing a first test of a suite of tests may be provided as input to a trained machine learning model(s) to classify one or more tests of the suite of tests as having a particular characteristic. A to-be-executed test may be classified as likely to pass or likely to fail when executed, for example. An already-executed test may be classified as reliable or unreliable, as another example. Based on the classification of the test(s), the suite of tests may be modified to optimize testing of the wireless carrier's system.

Abstract: The techniques described herein involve analysis of communication data included in trace file(s) of device(s) involved in a communication. These trace file(s) may each include data associated with multiple layers of a communication protocol stack of a respective device or data associated with a single such layer. The techniques may further involve one or more of determination of performance metrics associated with data at a specific layer of a specific device, correlation of the data between layers of a device, or correlation of data across multiple device(s) involved in the communication. The performance metrics or correlated data may then be analyzed based on thresholds or models to determine whether the performance metrics or correlated data exhibits a degraded quality of user experience. Also or instead, graphic or textual representations of the performance metrics or correlated data may be generated.

Abstract: In a software testing environment, a test script may be designed to search among outputs of an in-test target application for an expected output element having a specified property. A test execution engine executes the test script, and if such an output is not found, and the test script returns a “fail” result, the test execution engine revises the test script so that rather than searching for the originally specified property, the test script searches for a different property, where the different property is a property that the expected output element was observed to have during a previous execution of the test script. The test execution engine then executes the revised test script and reports its results.

Abstract: Systems and methods for testing power consumption in user equipment (UE). The system can include devices detecting changes in power consumption due to updates in hardware, software, or both. The system can test multiple UEs to create baseline consumption measurements to detect anomalies between UEs. The system can include individual power monitoring for each UE in a “UE farm.” The system can use a workstation and a microcontroller to manage multiple UEs to perform tasks at the same time or at predetermined intervals. The system can provide benchmark testing for UEs to identify anomalies. The method can instruct multiple UEs in the UE farm to perform a particular function, run a baseline set of applications, or test new applications to identify changes in consumption caused by applications, updates, or UEs. The systems and methods can provide a standardized means for rating power consumption on mobile electronic equipment.

Abstract: Systems and methods for remotely resetting networked devices is disclosed. The system can comprise a primary network connection and a secondary network connection to provide communications redundancy. The system can also include physical connections between networked devices to enable devices to be reset regardless of the type of error. The method can include monitoring one or more parameters of a device to determine what type of error is occurring and what action to take. The system can enable a first computer (or other electronic device) to restart applications, adapters, and services on a second computer (or other electronic device). The system can also enable the first computer to reset or restart the second computer. The system can enable networked computers to monitor and restore operation to other computers on the same network without human intervention.

Abstract: A system for testing a network-based application has a continuous integration (CI) service that performs CI testing of a server application that is being developed to support a client application, such as a client application that will be used on a wireless communication device. The CI service detects server source code changes and in response rebuilds the server application and deploys it to a test server. In addition, the client application is installed and executed on one or more wireless communication devices, so that the client application communicates with the test server using a wireless communications network. Tests of the client application are performed as it executes on the devices, and results are reported to the CI service. The CI service reports any errors that occurred during testing of either the server application or the client application.

Abstract: Systems and methods for remotely resetting networked devices is disclosed. The system can comprise a primary network connection and a secondary network connection to provide communications redundancy. The system can also include physical connections between networked devices to enable devices to be reset regardless of the type of error. The method can include monitoring one or more parameters of a device to determine what type of error is occurring and what action to take. The system can enable a first computer (or other electronic device) to restart applications, adapters, and services on a second computer (or other electronic device). The system can also enable the first computer to reset or restart the second computer. The system can enable networked computers to monitor and restore operation to other computers on the same network without human intervention.

Abstract: A cellular communications network is supported by an infrastructure that has many different parts. A fault management system can be used to monitor status and faults of the infrastructure. In response to certain types of faults, the fault management system initiates diagnostic testing, which may be performed using one or more communication devices such as those that might commonly be used in conjunction with the cellular communications network. One type of test might consist of installing an application on a communication device and measuring responsiveness of the application. The application is selected as one that relies at least in part on communicating through the cellular communications network. The responsiveness of the application may therefore be relevant to evaluating network performance and infrastructure status.

Abstract: Software applications are tested in different contexts, such as on different devices and under different conditions. During initial testing of an application, conditions of contexts are selected randomly, and the application is tested in each resulting context. After obtaining results from a sufficient number of contexts, the results are analyzed to create a predictive model indicating, for any postulated context, whether testing of the application is most likely to fail or to otherwise produce negative test results. The model is then analyzed to identify contexts that are most likely to produce negative results or failures, and those contexts are emphasized in subsequent application testing.

Abstract: The techniques described herein involve analysis of communication data included in trace file(s) of device(s) involved in a communication. These trace file(s) may each include data associated with multiple layers of a communication protocol stack of a respective device or data associated with a single such layer. The techniques may further involve one or more of determination of performance metrics associated with data at a specific layer of a specific device, correlation of the data between layers of a device, or correlation of data across multiple device(s) involved in the communication. The performance metrics or correlated data may then be analyzed based on thresholds or models to determine whether the performance metrics or correlated data exhibits a degraded quality of user experience. Also or instead, graphic or textual representations of the performance metrics or correlated data may be generated.

Abstract: Monitoring devices, which in some cases may comprise smartphones, are placed at base stations of a cellular communications network to report base station information. Each monitoring device is configured to periodically determine whether its associated base station is operating correctly, or whether there is some type of anomaly at the base station. As long as there is no anomaly, the monitoring device repeatedly sends status messages to a support service of the cellular communications network. Each such status message indicates that the monitoring device has performed an analysis and has determined there is no serious anomaly at the base station. The support service monitors status messages from multiple monitoring devices. If the support service does not receive a status message within a given period of time, the support service may alert service personnel and/or automatically dispatch support resources to the base station or trigger diagnostics and/or corrective measures.