Abstract: According to an aspect, a method may include executing, by a computing device, a television application. A method may include gathering, by the computing device, information and data related to interactions of a user with a user interface of the television application. A method may store the information and data related to the interactions of the user locally on the computing device. A method may generate an on-device model associated with the user based on the information and data related to the interactions of the user with the user interface of the television application. A method may determine media content recommendations for the user utilizing the on-device model. A method may integrate the media content recommendations in the user interface of the television application.

Abstract: A method may include displaying, by a computing device, media content recommendations in a user interface of a television application based on at least one known preference of a user. A method may receive, by the computing device, a request for media content suggestions not based on the at least one known preference of the user. A method may in response to receiving the request for the media content suggestions, initiate, by the television application, a stateless recommendation process. A method may generate, by the stateless recommendation process, the media content suggestions. A method may display the media content suggestions in the user interface of the television application, and may receive a selection of at least one media content item associated with a respective media content suggestion. A method may update, by the stateless recommendation process, the media content suggestions based on the at least one media content item.

Abstract: According to an aspect, a method may generate an activity timeline of a user based on data indicative of a search history, a watch history, and activity of the user with a television application. The method may generate a sports score using a sports model based on the activity timeline of the user, may generate a ranked list of sports related media content items for recommending to the user based on the sports score, and may associate a respective selectable information item with each of the sports related media content items included in the ranked list. The method may send, to a network-connected display device, the ranked list of sports related media content items that includes each respective selectable information item for displaying each respective selectable information item in a user interface of the television application executing on the network-connected display device.

Abstract: According to an aspect, a method may include receiving a request to generate an autoplay list, identifying a plurality of media content items, identifying a runtime for each of the plurality of media content items, based on determining that a runtime for a media content item included in the plurality of media content items exceeds a predefined runtime, generating a clip of the media content item with a runtime that is less than or equal to the predefined runtime, generating an ordered list of the plurality of media content items, substituting the clip for the media content item in place of the media content item included in the ordered list, generating an autoplay list based on the ordered list, and sending the autoplay list for a television application to play the plurality of media content items in the autoplay list on a network-connected display device.

Abstract: According to an aspect, a method may include receiving a request to generate an autoplay list, identifying a plurality of media content items, identifying a runtime for each of the plurality of media content items, based on determining that a runtime for a media content item included in the plurality of media content items exceeds a predefined runtime, generating a clip of the media content item with a runtime that is less than or equal to the predefined runtime, generating an ordered list of the plurality of media content items, substituting the clip for the media content item in place of the media content item included in the ordered list, generating an autoplay list based on the ordered list, and sending the autoplay list for a television application to play the plurality of media content items in the autoplay list on a network-connected display device.

Abstract: According to an aspect, a method may generate an activity timeline of a user based on data indicative of a search history, a watch history, and activity of the user with a television application. The method may generate a sports score using a sports model based on the activity timeline of the user, may generate a ranked list of sports related media content items for recommending to the user based on the sports score, and may associate a respective selectable information item with each of the sports related media content items included in the ranked list. The method may send, to a network-connected display device, the ranked list of sports related media content items that includes each respective selectable information item for displaying each respective selectable information item in a user interface of the television application executing on the network-connected display device.

Abstract: Systems, methods, and computer-readable media are disclosed for language-agnostic subtitle drift detection and correction. A method may include determining subtitles and/or captions from media content (e.g., videos), the subtitles and/or captions corresponding to dialog in the media content. The subtitles may be broken up into segments which may be analyzed to determine a likelihood of drift (e.g., a likelihood that the subtitles are out of synchronization with the dialog in the media content) for each segment. For segments with a high likelihood of drift, the subtitles may be incrementally adjusted to determine an adjustment that eliminates and/or reduces the amount of drift, and the drift in the segment may be corrected based on the drift amount detected. A linear regression model and/or human blocks determined by human operators may be used to otherwise optimize drift correction.

Abstract: Systems, methods, and computer-readable media are disclosed for language-agnostic subtitle drift detection and correction. A method may include determining subtitles and/or captions from media content (e.g., videos), the subtitles and/or captions corresponding to dialog in the media content. The subtitles may be broken up into segments which may be analyzed to determine a likelihood of drift (e.g., a likelihood that the subtitles are out of synchronization with the dialog in the media content) for each segment. For segments with a high likelihood of drift, the subtitles may be incrementally adjusted to determine an adjustment that eliminates and/or reduces the amount of drift, and the drift in the segment may be corrected based on the drift amount detected. A linear regression model and/or human blocks determined by human operators may be used to otherwise optimize drift correction.

Abstract: Systems, methods, and computer-readable media are disclosed for language-agnostic subtitle drift detection and correction. A method may include determining subtitles and/or captions from media content (e.g., videos), the subtitles and/or captions corresponding to dialog in the media content. The subtitles may be broken up into segments which may be analyzed to determine a likelihood of drift (e.g., a likelihood that the subtitles are out of synchronization with the dialog in the media content) for each segment. For segments with a high likelihood of drift, the subtitles may be incrementally adjusted to determine an adjustment that eliminates and/or reduces the amount of drift and the drift in the segment may be corrected based on the drift amount detected. A linear regression model and/or human blocks determined by human operators may be used to otherwise optimize drift correction.

Abstract: Devices, systems, and methods are provided for language-agnostic subtitle drift detection and localization. A method may include extracting audio from video, dividing the audio into overlapping blocks, and determining the probabilities of overlapping portions of the blocks, the probabilities indicating a presence of voice data represented by the audio in the blocks. The method may generate machine blocks using overlapping portions of blocks where voice data is present, and may map the machine blocks to corresponding blocks indicating that subtitles are available for the video. For mapped blocks, the method may include determining features such as when subtitles are available without voice audio, when voice audio is available without subtitles, and when voice audio and subtitles both are available. Using the features, the method may include determining the probability that the video includes subtitle drift, and the method may include analyzing the video to localize where the subtitle drift occurs.