Abstract: An insulin infusion set (IIS) device with one or more features designed to achieve longevity in a patients continuous subcutaneous insulin infusion (CSII) site viability. One exemplary feature is an adsorbent material configured to adsorb phenolic excipients (e.g., m-cresol or other preservatives) from the insulin formulation. The adsorbent material may be positioned along a fluid pathway specifically designed to increase and/or extend exposure between the insulin formulation and the adsorbent material. Another exemplary feature is a medicament configured to reduce the patients inflammation or slow the progression of the patients inflammatory response. Yet another exemplary feature is a diffusive catheter configured to deliver the insulin formulation in a diffuse manner.

Abstract: Embodiments of a method for making a contoured glass article and the resulting contoured glass article are disclosed. In one or more embodiments, the method includes cold bending a flat glass sheet having first and second opposing major surfaces, at least one region having a first thickness, and at least one region having a second thickness that is less than the first thickness, to produce cold bent glass sheet having at least one bend region along a portion of the at least one region having the second thickness; and restraining the cold bent glass sheet to produce the contoured glass article. One or more embodiments pertain to the resulting contoured glass article.

Abstract: Systems and methods are presented herein for recording portions of a media asset relevant to recording criteria. A media application receives input indicating the recording criteria and identifying a first keyword. The media application accesses a data structure to identify a first node associated with the first keyword. The data structure includes the first node and a plurality of nodes connected to the first node via a plurality of paths. The media application receiving audio component data for a portion of the media asset extracts a term from the audio component data, and identifies a second node in the data structure that is associated with the extracted term. The media application calculates a path score for the portion of the media asset based on a path size in the data structure between the first node and the second node. When the score is high enough, the portion of the media asset is recorded.

Abstract: Systems and methods are provided herein for automatically generating universal metadata for non-linear content when the non-linear content is determined to be viral. To this end, the systems and methods determine a first rate at which first users are consuming non-linear content from a first source. In response to determining that the first rate exceeds the threshold, the systems and methods determine a second source that from which second users are consuming the non-linear content, and a second rate at which the second users are consuming the non-linear content from the second source. In response to determining that the second rate exceeds the threshold, the systems and methods generate universal metadata for the non-linear content by automatically populating metadata fields of by scraping metadata fitting predefined criteria corresponding to the metadata fields from the first source and from the second source, and store the populated metadata fields to a database.

Abstract: An injection device assembly including a housing, a syringe, a drive mechanism and a status sensing system and a module are described. The drive mechanism advances the syringe from a storage position to an injection position, and a plunger advances the syringe piston from an initial position to a final position. The status sensing system includes a light emitter that emits electromagnetic radiation into the housing transverse to the longitudinal axis, a light detector and a controller. The system determines a current status condition from at least three of the following five status conditions: 1) an injection ready state; 2) a needle insertion state; 3) a drug delivered state; 4) a needle guard present state; and 5) a needle retraction state.

Abstract: A method for training an endpointer model includes short-form speech utterances and long-form speech utterances. The method also includes providing a short-form speech utterance as input to a shared neural network, the shared neural network configured to learn shared hidden representations suitable for both voice activity detection (VAD) and end-of-query (EOQ) detection. The method also includes generating, using a VAD classifier, a sequence of predicted VAD labels and determining a VAD loss by comparing the sequence of predicted VAD labels to a corresponding sequence of reference VAD labels. The method also includes, generating, using an EOQ classifier, a sequence of predicted EOQ labels and determining an EOQ loss by comparing the sequence of predicted EOQ labels to a corresponding sequence of reference EOQ labels. The method also includes training, using a cross-entropy criterion, the endpointer model based on the VAD loss and the EOQ loss.

Abstract: A system for generating pressure pulses in drilling fluid includes a concentric drive power section. The power section includes a stator and a rotor rotatably disposed in the stator. The rotor is coaxially aligned with the stator. The system also includes a valve. The valve includes a first valve member coupled to the stator and a second valve member coupled to the rotor. The second valve member is configured to rotate with the rotor relative to the first valve member and the stator. The rotation of the second valve member relative to the first valve member is configured to generate pressure pulses in drilling fluid flowing through the concentric drive power section.

Abstract: Systems and methods are described herein for controlling playback and/or recording of media assets at a media device based on the state of a secondary device. Specifically, while displaying a media asset, a media device receives a signal from the secondary device indicating a change of state of the secondary device. The media device displays a notification of the signal, and the user can interact with the notification. If the user interacts with the notification for more than a threshold amount of time, the media device may begin recording the media asset. Once the user has finished interacting with the notification, the media device may stop recording and the user can see content that they may have missed by reviewing the recorded portion of the media asset.

Abstract: Systems and methods are disclosed herein for enabling multiple users' digital rights to be applied as needed to access media while those users are present to consume media. In particular, a media guidance application may receive a selection of a media asset and determine that the user equipment device does not allow access to the media asset. In response, the media guidance application may identify all the users consuming content from that user equipment device and retrieve digital rights for each user. The media guidance application may compare the digital rights of each user with digital rights required to access the media asset. The media guidance application may, based on the comparison, identify a set of digital rights (e.g., for another user present) that enable access to the media asset, and use those digital rights to access the media asset for consumption.

Abstract: Systems and methods are provided herein for identifying viewers who are engaged in a media asset but unlikely to consume secondary content, and presenting, to the identified viewers, some secondary content designed to incentivize the identified viewers to consume secondary content. These systems and methods determine that a viewer is engaged in a media asset upon determining level of engagement of the viewer in the media asset is greater than a threshold value. These systems and methods determine that the viewer is unlikely to consume secondary content upon determining level of engagement of the viewer in secondary content is less than a threshold value. These systems and methods, upon determining that the viewer is engaged in the media asset but unlikely to consume secondary content, generate for display secondary content related to the media asset during breaks in the media asset.

Abstract: Systems and methods are presented herein for recording portions of a media asset relevant to recording criteria. A media application receives input indicating the recording criteria and identifying a first keyword. The media application accesses a data structure to identify a first node associated with the first keyword. The data structure includes the first node and a plurality of nodes connected to the first node via a plurality of paths. The media application receiving audio component data for a portion of the media asset extracts a term from the audio component data, and identifies a second node in the data structure that is associated with the extracted term. The media application calculates a path score for the portion of the media asset based on a path size in the data structure between the first node and the second node. When the score is high enough, the portion of the media asset is recorded.

Abstract: Systems and methods are disclosed herein for identifying users based on voice data and media consumption data. A media guidance application may generate a voice signature from the user's input and based on that signature identify some demographic characteristics of the user (e.g., age, gender, and other suitable characteristics). The media guidance application may retrieve user data for users that are associated with a household of the user and attempt to identify which of the users spoke the command. If multiple users are identified, based on the demographic characteristics, the media guidance application may use the content of the voice command (e.g., a type of media requested) to identify the user.

Abstract: A method for training an endpointer model includes short-form speech utterances and long-form speech utterances. The method also includes providing a short-form speech utterance as input to a shared neural network, the shared neural network configured to learn shared hidden representations suitable for both voice activity detection (VAD) and end-of-query (EOQ) detection. The method also includes generating, using a VAD classifier, a sequence of predicted VAD labels and determining a VAD loss by comparing the sequence of predicted VAD labels to a corresponding sequence of reference VAD labels. The method also includes, generating, using an EOQ classifier, a sequence of predicted EOQ labels and determining an EOQ loss by comparing the sequence of predicted EOQ labels to a corresponding sequence of reference EOQ labels. The method also includes training, using a cross-entropy criterion, the endpointer model based on the VAD loss and the EOQ loss.

Abstract: Systems and methods are provided herein for automatically generating universal metadata for non-linear content when the non-linear content is determined to be viral. To this end, the systems and methods determine a first rate at which first users are consuming non-linear content from a first source. In response to determining that the first rate exceeds the threshold, the systems and methods determine a second source that from which second users are consuming the non-linear content, and a second rate at which the second users are consuming the non-linear content from the second source. In response to determining that the second rate exceeds the threshold, the systems and methods generate universal metadata for the non-linear content by automatically populating metadata fields of by scraping metadata fitting predefined criteria corresponding to the metadata fields from the first source and from the second source, and store the populated metadata fields to a database.

Abstract: A system for generating an output audio signal includes a context encoder, a text-prediction network, and a text-to-speech (TTS) model. The context encoder is configured to receive one or more context features associated with current input text and process the one or more context features to generate a context embedding associated with the current input text. The text-prediction network is configured to process the current input text and the context embedding to predict, as output, a style embedding for the current input text. The style embedding specifies a specific prosody and/or style for synthesizing the current input text into expressive speech The TTS model is configured to process the current input text and the style embedding to generate an output audio signal of expressive speech of the current input text. The output audio signal has the specific prosody and/or style specified by the style embedding.

Abstract: Aspects of the disclosure relate to packagable RNA constructs with minimal intervening viral sequences. These constructs can be used to generate lentiviral viruses encoding large genes capable of transducing primary human cells.

Abstract: Systems and methods are provided for intelligently accessing media content based on rights. This may be accomplished by a media guidance application that generates a DRM score associated with a first copy of the media asset. The media guidance application determines that plurality of alternative copies of a media asset are available from other sources and determines DRM scores for the alternative copies. The media guidance application determines that one of the alternative copies has a lower digital rights management score as compared to the first copy of the media asset and stores the alternative copy instead of the first copy of the media asset.

Abstract: Various embodiments are related to substrates having one or more well structures with a trapezoidal cylinder shaped through hole via extending from the bottom of the well structure though the substrate.

Abstract: A method of performing ion exchange of a thin, flexible glass substrate having an average thickness equal to or less than about 0.3 mm to chemically strengthen the glass substrate is disclosed. The chemically strengthened glass substrate comprises a first compressive stress layer having a first depth of layer, and a second compressive stress layer having a second depth of layer, the first and second stress layers being separated by a layer of tensile stress. A laminated article comprising the chemically strengthened glass substrate is also described.

Abstract: Systems and methods are disclosed herein for identifying users based on voice data and media consumption data. A media guidance application may generate a voice signature from the user's input and based on that signature identify some demographic characteristics of the user (e.g., age, gender, and other suitable characteristics). The media guidance application may retrieve user data for users that are associated with a household of the user and attempt to identify which of the users spoke the command. If multiple users are identified, based on the demographic characteristics, the media guidance application may use the content of the voice command (e.g., a type of media requested) to identify the user.