Abstract: A computer-implemented method of generating speech audio in a video game is provided. The method includes inputting, into a synthesizer module, input data that represents speech content. Source acoustic features for the speech content in the voice of a source speaker are generated and are input, along with a speaker embedding associated with a player of the video game into an acoustic feature encoder of a voice convertor. One or more acoustic feature encodings are generated as output of the acoustic feature encoder, which are inputted into an acoustic feature decoder of the voice convertor to generate target acoustic features. The target acoustic features are processed with one or more modules, to generate speech audio in the voice of the player.

Abstract: This specification describes a computer-implemented method of generating speech audio for use in a video game, wherein the speech audio is generated using a voice convertor that has been trained to convert audio data for a source speaker into audio data for a target speaker. The method comprises receiving: (i) source speech audio, and (ii) a target speaker identifier. The source speech audio comprises speech content in the voice of a source speaker. Source acoustic features are determined for the source speech audio. A target speaker embedding associated with the target speaker identifier is generated as output of a speaker encoder of the voice convertor. The target speaker embedding and the source acoustic features are inputted into an acoustic feature encoder of the voice convertor. One or more acoustic feature encodings are generated as output of the acoustic feature encoder. The one or more acoustic feature encodings are derived from the target speaker embedding and the source acoustic features.

Abstract: Disclosed are systems and methods for profiling a plurality of companies. The companies are profiled by receiving HTML files on the world wide web that contain hyperlinks to a domain name of one or more of the plurality of companies; determining an ingress of each of the plurality of companies based on a number of hyperlinks to the domain name of that company in the HTML files; receiving industry categories and industry embedding values for each of the plurality of companies; and designating a first company and a second company of the plurality of companies as similar based at least in part on one or more of the ingress of the first company, the ingress of the second company, a semantic distance between the industry embedding values of the first company and the industry embedding values of the second company, and a number of industry categories common between the first company and the second company.

Abstract: A system for use in video game development to generate expressive speech audio comprises a user interface configured to receive user-input text data and a user selection of a speech style. The system includes a machine-learned synthesizer comprising a text encoder, a speech style encoder and a decoder. The machine-learned synthesizer is configured to generate one or more text encodings derived from the user-input text data, using the text encoder of the machine-learned synthesizer; generate a speech style encoding by processing a set of speech style features associated with the selected speech style using the speech style encoder of the machine-learned synthesizer; combine the one or more text encodings and the speech style encoding to generate one or more combined encodings; and decode the one or more combined encodings with the decoder of the machine-learned synthesizer to generate predicted acoustic features.

Abstract: A system for use in video game development to generate expressive speech audio comprises a user interface configured to receive user-input text data and a user selection of a speech style. The system includes a machine-learned synthesizer comprising a text encoder, a speech style encoder and a decoder. The machine-learned synthesizer is configured to generate one or more text encodings derived from the user-input text data, using the text encoder of the machine-learned synthesizer; generate a speech style encoding by processing a set of speech style features associated with the selected speech style using the speech style encoder of the machine-learned synthesizer; combine the one or more text encodings and the speech style encoding to generate one or more combined encodings; and decode the one or more combined encodings with the decoder of the machine-learned synthesizer to generate predicted acoustic features.

Abstract: A hybrid prediction system may aggregate electronic data to identify and initially predict an outcome of a future event and subsequently update the initial prediction. The system may include at least one processor and a memory. The processor may access data scraped from the Internet. The data may be associated with at least one future event. The processor may further store the scraped data, determine, from the scraped data, an initial prediction of the outcome of the at least one future event, generate, from the scraped data, an initial likelihood indication associated with the initial prediction, and transmit the initial prediction and the initial likelihood indication to a device associated with one or more users. The processor may further receive proprietary information, store the proprietary information, determine, using the scraped data and the proprietary information, a subsequent likelihood indication, and transmit the subsequent likelihood indication to the device.

Abstract: A system for use in video game development to generate expressive speech audio comprises a user interface configured to receive user-input text data and a user selection of a speech style. The system includes a machine-learned synthesizer comprising a text encoder, a speech style encoder and a decoder. The machine-learned synthesizer is configured to generate one or more text encodings derived from the user-input text data, using the text encoder of the machine-learned synthesizer; generate a speech style encoding by processing a set of speech style features associated with the selected speech style using the speech style encoder of the machine-learned synthesizer; combine the one or more text encodings and the speech style encoding to generate one or more combined encodings; and decode the one or more combined encodings with the decoder of the machine-learned synthesizer to generate predicted acoustic features.

Abstract: Using user-specific prediction models, it is possible to present an individualized view of messages generated by users playing a shared instance of a video game. Further, users with different subjective views of what is offensive may be presented with different forms or annotations of a message. By personalizing the views of messages generated by users, it is possible to reduce or eliminate the toxic environment that sometimes forms when players, who may be strangers to each other and may be located in disparate locations play a shared instance of a video game. Further, the user-specific prediction models may be adapted to filter or otherwise annotate other undesirable messages that may not be offensive, such as a message generated by one user in a video game that includes a solution to an in-game puzzle that another user may not desire to read as it may spoil the challenge for the user.

Abstract: A text analytics system may ascertain sentiment about multi-sectioned documents and may associate the sentiment with particular sections. The system may include at least one processor configured to scrape the Internet for text data associated with comments expressed by a plurality of individuals about a common multi-sectioned document. The comments may not be not linked to a particular section. The at least one processor may be further configured to analyze the text data in order to determine a sentiment associated with each comment; apply an association analysis filter to the text data in order to correlate at least a portion of each comment with one or more sections of the multi-sectioned document; and transmit for display to the system user a visualization of the sentiment mapped to one or more sections of the multi-sectioned document.

Abstract: Using user-specific prediction models, it is possible to present an individualized view of messages generated by users playing a shared instance of a video game. Further, users with different subjective views of what is offensive may be presented with different forms or annotations of a message. By personalizing the views of messages generated by users, it is possible to reduce or eliminate the toxic environment that sometimes forms when players, who may be strangers to each other and may be located in disparate locations play a shared instance of a video game. Further, the user-specific prediction models may be adapted to filter or otherwise annotate other undesirable messages that may not be offensive, such as a message generated by one user in a video game that includes a solution to an in-game puzzle that another user may not desire to read as it may spoil the challenge for the user.

Abstract: Systems and methods are disclosed herein for using machine learning to automatically modify unstructured scripts with speech tags for a context in which the speech is to be spoken so that the speech can be synthesized to sound more realistic and more contextually appropriate. The systems and methods can be dynamically applied. Training context tags and corresponding structured training scripts are used to train the machine learning system to generate an AI model. The AI model can be used in different ways, and a feedback system is described.

Abstract: A collaborative prediction system for altering predictive outcomes of dynamic processes may include a processor configured to store initial information about a specific dynamic process having a plurality of potentially differing outcomes, assign to the specific dynamic process a first likelihood of occurrence of at least one of the potentially differing outcomes, and receive, from a first system user, notification data, the notification data being associated with the specific dynamic process.

Abstract: A system for normalizing aggregated electronic data may include at least one processor and a memory. The processor may access data scraped from a plurality of local databases, each storing localized terms for temporal local milestones; for example, a first localized term for a first temporal milestone in a first locale may differ from a second localized term for a similar second temporal milestone in a second locale. The processor may store the temporal local milestones in a central database, determine a mapping of each of the localized terms for each temporal local milestone to a normalized term for each milestone, and identify, in a displayed timeline associated with the first locale, each associated temporal local milestone using the normalized term for each milestone. The processor may further identify, in a displayed timeline associated with the second locale, each associated temporal local milestone using the normalized term for each milestone.

Abstract: An Internet-based agenda data analysis system may include at least one processor configured to maintain a list of user-selectable agenda issues, present to a user via a user interface, the list of user-selectable agenda issues, and receive via the user interface, based on a selection from the list, agenda issues of interest to an organization. The processor may be configured to access information scraped from the Internet to determine, for a plurality of policymakers, individual policymaker data from which an alignment position of each policymaker on each of the agenda issues is determinable, calculate alignment position data from the individual policymaker data, the alignment position data corresponding to relative positions of each of the plurality of policymakers on each of the plurality of selected issues, and transform the alignment position data into a graphical display that presents the alignment positions of multiple policymakers.

Abstract: A system for predicting and prescribing actions for impacting policymaking outcomes may include at least one processor configured to access first information scraped from the Internet to identify, for a particular pending policy, information about a plurality of policymakers slated to make a determination on the pending policy. The processor may parse the scraped first information to determine an initial prediction relating to an outcome of the pending policy. The processor may access second information to identify an action likely to change at least one of the initial prediction and the propensity of at least one policymaker, to thereby generate a subsequent prediction corresponding to an increase in a likelihood of achieving the desired outcome. The processor may display to the system user a recommendation to take the action in order to increase the likelihood of achieving the desired outcome.

Abstract: A text analytics system may predict whether a policy will be adopted. The system may include at least one processor configured to access information scraped from the Internet to identify text data associated with comments expressed by a plurality of individuals about a proposed policy. The at least one processor may be further configured to analyze the text data in order to determine a sentiment of each comment; apply an influence filter to each comment to determine an influence metric associated with each comment; weight each comment using the influence metric; determine based on an aggregate of the weighted comments, an indicator associated with adoption of the policy; and transmit the indicator to a system user.

Abstract: A text analytics system may ascertain sentiment about multi-sectioned documents and may associate the sentiment with particular sections. The system may include at least one processor configured to scrape the Internet for text data associated with comments expressed by a plurality of individuals about a common multi-sectioned document. The comments may not be not linked to a particular section. The at least one processor may be further configured to analyze the text data in order to determine a sentiment associated with each comment; apply an association analysis filter to the text data in order to correlate at least a portion of each comment with one or more sections of the multi-sectioned document; and transmit for display to the system user a visualization of the sentiment mapped to one or more sections of the multi-sectioned document.

Abstract: A collaborative prediction system for altering predictive outcomes of dynamic processes may include a processor configured to store initial information about a specific dynamic process having a plurality of potentially differing outcomes, assign to the specific dynamic process a first likelihood of occurrence of at least one of the potentially differing outcomes, and receive, from a first system user, notification data, the notification data being associated with the specific dynamic process.

Abstract: A hybrid prediction system may aggregate electronic data to identify and initially predict an outcome of a future event and subsequently update the initial prediction. The system may include at least one processor and a memory. The processor may access data scraped from the Internet. The data may be associated with at least one future event. The processor may further store the scraped data, determine, from the scraped data, an initial prediction of the outcome of the at least one future event, generate, from the scraped data, an initial likelihood indication associated with the initial prediction, and transmit the initial prediction and the initial likelihood indication to a device associated with one or more users. The processor may further receive proprietary information, store the proprietary information, determine, using the scraped data and the proprietary information, a subsequent likelihood indication, and transmit the subsequent likelihood indication to the device.

Abstract: A data pattern analysis system may determine patterns in aggregated electronic data from at least one Internet server. The system may include at least one processor configured to access data scraped from the Internet to identify information that identifies individuals with a role in an outcome of at least one prospective future event related to a predefined subject matter, the information including portions unrelated to the subject matter of the at least one prospective future event, process the information to determine patterns in portions of the data unrelated to the predefined subject matter, determine an influence factor for each of the identified individuals, based, at least in part, on the patterns in the data unrelated to the predefined subject matter, and predict the outcome of the future event based at least in part on the patterns in the data unrelated to the predefined subject matter.