Abstract: The disclosed computer-implemented method may include (1) receiving a media stream, (2) determining a first priority level for the media stream, (3) generating one or more first encoded segments from a first segment of the media stream based on the first priority level, (4) determining, while generating the one or more first encoded segments, a second priority level for the media stream, the second priority level being different than the first priority level, (5) generating one or more second encoded segments from a second segment of the media stream based on the second priority level, and (6) streaming the one or more first encoded segments and the one or more second encoded segments to one or more client devices. Various other methods, systems, and computer-readable media are also disclosed.

Abstract: In one embodiment, a method includes sending, to a client system of a first user, a notification for presentation to the first user. The notification may include a reference to a media-content item and an activatable element for accessing the media-content item. A user input selecting the activatable element may be received. Responsive to the user input, a list of proximate media-player devices to the client system may be determined. The list of proximate media-player devices may be sent to the client system for presentation to the first user. Each of the proximate media-player devices may correspond to a deep link being selectable by the first user. Each deep link may include instructions for presenting the media-content item on a display device coupled to the respective media-player device.

Abstract: A method of generating respective adaptions of digital content for display by multiple computing devices is recited. In one embodiment, a first computing device sends an invitation to congregate to one or more second computing devices. If the first computing device determines that the one or more second computing devices are in physical proximity to the first computing device, the first computing device may invoke play of the digital content. A plurality of generated content adaptions may be generated, wherein a first content adaption is configured for synchronous display by the first computing device with one or more second content adaptions displayed on the one or more second computing devices, respectively.

Abstract: A client device includes a display device and one or more image capture devices configured to capture video data of a local area surrounding the client device. Different users of an online system are identified from the captured video data, provided privacy settings maintained for the users authorize their identification from the video data, and the client device transmits a request for content to the online system that identifies each of the identified users. The client device receives content items selected for each of the identified users from the online system. From the received content items, the client device generates a feed of content that includes at least one content item selected for each of the identified users and presents the generated feed of content via the display device.

Abstract: A computer-implemented method for automatically switching between media streams may include (i) identifying a plurality of media streams of an event, where one media stream from the plurality is designated as a primary stream of the event that is transmitted by default to at least one viewer of the event, (ii) selecting a media stream from the plurality of media streams to be the primary stream based on a favorable characteristic of the media stream, (iii) detecting that a different media stream within the plurality of media streams has a more favorable characteristic compared to the media stream, (v) in response to detecting that the different media stream has the more favorable characteristic, automatically selecting the different media stream to be the primary stream and (v) transmitting the different media stream to the at least one viewer of the event. Various other methods, systems, and computer-readable media are also disclosed.

Abstract: In one embodiment, a method includes receiving, at a first client system of a first user, a first notification that a second user has performed a first in-game action in a game session of a game; displaying, at the first client system, a game window associated with the game within a message thread on a messaging application running on the first client system, the game window comprising a live gameplay of the first in-game action and one or more interactive action buttons operable to allow the first user to perform one or more in-game actions; receiving, by the first client system, an indication of the first user performing a second in-game action via the one or more interactive action buttons; and sending, by the first client system to a second client system, a second notification that the first user has performed the second in-game action in the game session.

Abstract: Systems and methods for digitally fetching music content may include (1) providing a user of a device with a music service accessible via the device, (2) receiving a digital user request to (i) select a music composition based on content from an additional music service and (ii) digitally fetch the music composition for use via the music service, and, in response to receiving the digital user request, (3) analyzing the content and selecting the music composition based on the analysis of the content, and (4) digitally fetching the music composition for use via the music service. Various other methods, systems, and computer-readable media are also disclosed.

Abstract: A system comprises a data input vector unit, a weight input vector unit, and a plurality of calculation units of a matrix processor unit. The data input vector unit is configured to concurrently receive elements of different rows of a first and second data matrix. The weight input vector unit is configured to receive a combined weight vector and at least in part concurrently provide obtained weight elements of a first and second weight matrix to a corresponding first and second group of calculation units. Each calculation unit of the first and second group of calculation units is configured to multiply elements from the data input vector unit with elements of the corresponding weight matrix from the weight input vector unit and sum together multiplication results of the corresponding calculation unit to at least in part determine a corresponding element in a first or second convolution result matrix.

Abstract: Systems and methods for integrated audioconferencing social media music streaming may include (1) detecting that a user of a messaging application, which enables audioconferencing between its users, has selected a drop-in mode that enables audioconferencing between (i) the user and (ii) other users with access to a drop-in link corresponding to the user, (2) providing the drop-in link to contacts of the user in a messaging interface of the messaging application, and (3) in response to determining that a contact of the user has selected the drop-in link, automatically initiating an audioconference between the user and the contact. Various other methods, systems, and computer-readable media are also disclosed.

Abstract: One or more write efficiency metrics are determined. The one or more write efficiency metrics are associated with a tape storage drive while the tape storage drive is operating on one or more tape storage media cartridges to write data sets to the one or more tape storage media cartridges. It is predicted whether the tape storage drive would be able to write an entire set of data sets to a current tape storage media cartridge based on the one or more write efficiency metrics. It is determined whether a different tape storage drive is to be utilized to complete writing of the entire set of data sets to the current tape storage media cartridge based on the prediction.

Abstract: A computer-implemented method for translating with limited attention a sequence of input tokens into a sequence of output tokens may include (1) encoding by an encoder recurrent neural network (RNN) N input tokens of a sequence of (N+D) input tokens into an attention matrix limited to N input tokens, (2) iteratively updating a previous version of the attention matrix with the context vector generated from each excess input token yielding a final attention matrix at the last excess input token (D), and (3) decoding an output token using a decoder RNN with inputs from at least the weighted context vector and the just previously-determined output token of the decoder. The weighted context vector may be derived from an attention RNN processing the elements of the final attention matrix for each of the decoder output tokens. Various other methods, systems, and computer-readable media are also disclosed.

Abstract: A user-activity video signal can be received. The user-activity video signal can comprise a video capture characterizing a user's interaction with a software application on a client. The software application can have an interface display space on the client and the software application can execute on the client in operation with at least one remote backend computing system. Backend user interaction data corresponding to the user's interaction with the software application as generated by the at least one remote backend computing system can be received. A new video comprising a plurality of divided segments of the user-activity video signal can be generated automatically using the backend user interaction data and a plurality of predefined rules. The new video can be provided. Related apparatus, systems, techniques, and articles are also described.

Abstract: In one embodiment, a technique includes receiving a first story fragment from a storyteller interface of a first computing system. The technique further includes displaying the first story fragment on an audience interface of a second computing system. The technique also includes detecting a trigger on the audience interface of the second computing system. In the technique, the trigger corresponds to the first story fragment. The technique also includes identifying a special effect associated with the trigger. The technique further includes outputting the special effect and the first story fragment on at least one of the storyteller interface displayed on the first computing system and the audience interface displayed on the second computing system.

Abstract: Systems, methods, and non-transitory computer readable media are configured to provide a first element in an interface presentable to a user through which a plurality of ephemeral media content items are accessible by the user for a selected period of time. A second element in the interface through which a plurality of non-ephemeral media content items are accessible by the user is provided.

Abstract: An online system matches a user across multiple online systems based on image data for the user (e.g., profile photo) regardless whether the image data is from the online system, a different but related online system or a third party system. For example, to match the user across a social networking system and INSTAGRAM™ system, the online system compares the similarity between images of the user from both systems in addition to similarity of textual information in the user profiles on both systems. The similarity of image data and the similarity of textual information associated with the user are used by the online system as indicators of matched user accounts belonging to the same user across both systems. The online system applies models trained using deep learning techniques to match a user across multiple online systems based on the image data and textual information associated with the user.

Abstract: Systems, methods, and non-transitory computer-readable media can, for a first content item comprising a plurality of scan versions, calculate render scores for at least some of the plurality of scan versions, each render score being indicative of a likelihood of an associated scan version to render successfully on a first client computing device. A first scan version of the plurality of scan versions is selected based on the render scores. The first scan version of the first content item is transmitted to the first client computing device.

Abstract: Systems, methods, and non-transitory computer-readable media can generate individual feature data for each user of a plurality of users. A first cohort comprising a first plurality of users is generated, wherein the first plurality of users are selected from the plurality of users based on the individual feature data. A first set of cohort feature data associated with the first cohort is generated based on individual feature data for the first plurality of users. The first set of cohort feature data and a first set of cohort membership information are transmitted to a modeler. The first set of cohort membership information identifies each user of the plurality of users.

Abstract: One or more embodiments of the disclosure provide systems and methods for providing media presentations to users of a media presentation system. A media presentation generally includes a one or more media streams provided by one or more capturing users of the media presentation system. In one or more embodiments, a user of the media presentation system may share a media presentation with a viewing user. The media presentation system can provide a number of features to produce a media stream within a media presentation.