Abstract: The present invention relates to sulfonylureas and sulfonylthioureas comprising a 5-membered heteroaryl group substituted with a monovalent heteroaryl-containing group. The present invention further relates to salts, solvates and prodrugs of such compounds, to pharmaceutical compositions comprising such compounds, and to the use of such compounds in the treatment and prevention of medical disorders and diseases, most especially by the inhibition of NLRP3.

Abstract: The present disclosure provides an article of personal protective equipment (PPE). The article of PPE includes at least one sensor configured to generate motion signals indicative of a motion of a user of the article of PPE. The article of PPE further includes a controller configured to determine a plurality of motions of the user over a plurality of time periods. Each motion includes a corresponding magnitude and a corresponding direction over the corresponding time period. The controller is further configured to determine a motion pattern based on the plurality of motions, such that the motions pattern includes at least one motion from the plurality of motions that has a non-zero value for the corresponding magnitude. The controller is further configured to determine an emergency event at least based on the motion pattern. The controller is further configured to generate an alert signal upon determination of the emergency event.

Abstract: Systems and methods for text-to-speech with novel speakers can obtain text data and output audio data. The input text data may be input along with one or more speaker preferences. The speaker preferences can include speaker characteristics. The speaker preferences can be processed by a machine-learned model conditioned on a learned prior distribution to determine a speaker embedding. The speaker embedding can then be processed with the text data to generate an output that includes audio data descriptive of the text data spoken by a novel speaker.

Abstract: A method for estimating an embedding capacity includes receiving, at a deterministic reference encoder, a reference audio signal, and determining a reference embedding corresponding to the reference audio signal, the reference embedding having a corresponding embedding dimensionality. The method also includes measuring a first reconstruction loss as a function of the corresponding embedding dimensionality of the reference embedding and obtaining a variational embedding from a variational posterior. The variational embedding has a corresponding embedding dimensionality and a specified capacity. The method also includes measuring a second reconstruction loss as a function of the corresponding embedding dimensionality of the variational embedding and estimating a capacity of the reference embedding by comparing the first measured reconstruction loss for the reference embedding relative to the second measured reconstruction loss for the variational embedding having the specified capacity.

Abstract: A self-contained breathing apparatus (SCBA) facemask with a mask-mounted regulator having a pneumatic vibrating alert device. The pneumatic vibrating alert device is vibrationally coupled to at least one face-contacting component of the facemask and is acoustically isolated from the surrounding environment.

Abstract: Systems and methods for text-to-speech with novel speakers can obtain text data and output audio data. The input text data may be input along with one or more speaker preferences. The speaker preferences can include speaker characteristics. The speaker preferences can be processed by a machine-learned model conditioned on a learned prior distribution to determine a speaker embedding. The speaker embedding can then be processed with the text data to generate an output that includes audio data descriptive of the text data spoken by a novel speaker.

Abstract: A safety system for a user includes an article of personal protective equipment (PPE) configured to be at least partially in contact with the user. The safety system further includes at least one tactile actuator and a controller. The at least one tactile actuator is associated with the article of PPE. The at least one tactile actuator is configured to provide tactile stimulation to the user. The controller is communicably coupled to the at least one tactile actuator. The controller is configured to provide an actuating signal to the at least one tactile actuator to actuate the at least one tactile actuator based upon at least one predetermined condition. The at least one tactile actuator generates a tactile indication upon receiving the actuating signal.

Abstract: A method for estimating an embedding capacity includes receiving, at a deterministic reference encoder, a reference audio signal, and determining a reference embedding corresponding to the reference audio signal, the reference embedding having a corresponding embedding dimensionality. The method also includes measuring a first reconstruction loss as a function of the corresponding embedding dimensionality of the reference embedding and obtaining a variational embedding from a variational posterior. The variational embedding has a corresponding embedding dimensionality and a specified capacity. The method also includes measuring a second reconstruction loss as a function of the corresponding embedding dimensionality of the variational embedding and estimating a capacity of the reference embedding by comparing the first measured reconstruction loss for the reference embedding relative to the second measured reconstruction loss for the variational embedding having the specified capacity.

Abstract: A number of user interface devices are described. In one example, a user interface device includes a lens having a graphic visible through the lens, a transparent circuit film comprising a top surface and a bottom surface, a reflector film bonded to the bottom surface of the transparent circuit film, the reflector film including an embossed area which defines a pocket between the reflector film and the bottom surface of the transparent circuit film, a light emitting diode (LED) bonded to the bottom surface of the transparent circuit film and positioned within the pocket between the reflector film and the bottom surface of the transparent circuit film, and a layer of transparent pressure-sensitive adhesive interposed between the bottom surface of the lens and the top surface of the transparent circuit film.

Abstract: Methods, systems, and apparatus for performing speech recognition. In some implementations, acoustic data representing an utterance is obtained. The acoustic data corresponds to time steps in a series of time steps. One or more computers process scores indicative of the acoustic data using a recurrent neural network to generate a sequence of outputs. The sequence of outputs indicates a likely output label from among a predetermined set of output labels. The predetermined set of output labels includes output labels that respectively correspond to different linguistic units and to a placeholder label that does not represent a classification of acoustic data. The recurrent neural network is configured to use an output label indicated for a previous time step to determine an output label for the current time step. The generated sequence of outputs is processed to generate a transcription of the utterance, and the transcription of the utterance is provided.

Abstract: A safety system includes a plurality of alert devices associated with a unique identifier. Each alert device includes a user interface, a wireless transceiver, and a processor communicably coupled to the wireless transceiver. The processor of at least one alert device is configured to generate an alarm signal indicative of an alarm state, and transmit the alarm signal including the unique identifier of the at least one alert device to the wireless transceivers of the other alert devices. The processor of each of the other alert devices is configured to receive the alarm signal from the at least one alert device, determine a link quality indicative of at least a relative distance between the at least one alert device and the corresponding other alert device, and display, via the user interface, an alert message including the unique identifier of the at least one alert device and the corresponding link quality.

Abstract: A system, wearable device and management device provided for predicting a flashover event. According to one aspect of the disclosure, a wearable device for predicting a flashover event is provided. The wearable device includes at least one thermal sensor configured to generate thermal data associated with an environment, and processing circuitry configured to: determine a risk of ignition of at least one combustible gas in the environment based on the thermal data, and trigger at least one action based on the determined risk of ignition.

Abstract: A safety circuit for a Personal Alert Safety System (PASS) device includes a safety logic circuit and an alarm circuit. The safety logic circuit is communicably coupled to the PASS device. The safety logic circuit includes a timer configured to determine a time elapsed since the timer has been last reset. The safety logic circuit is configured to switch from an off state to an on state upon receiving an activation signal from the PASS device and reset the timer upon switching to the on state, generate a trigger signal in response to the timer exceeding a time threshold, and reset the timer upon receiving a reset signal from the PASS device before the timer exceeds the time threshold. The alarm circuit is communicably coupled to the safety logic circuit and configured to generate an alarm signal upon receiving the trigger signal from the safety logic circuit.

Abstract: A monitoring unit for wireless communication with a plurality of self-contained breathing apparatus (SCBA) units configured to form a network is provided. The monitoring unit includes a communication interface and processing circuitry. The processing circuitry is configured to receive, via the communication interface, a plurality of updates from each of the SCBA units where each update includes information indicating at least one communication link quality to at least one other SCBA unit. The processing circuitry is further configured to determine at least one distance from each SCBA unit to at least one other SCBA unit that is one hop away in the network based at least in part on the communication link quality. The processing circuitry is further configured to cause the determined at least one distance to be displayed.

Abstract: A tracker device for a first responder including a radio configured to receive a first radio signal from an emitter, and receive a second radio signal from an unmanned aerial vehicle (UAV). The second radio signal includes at least one data packet including location information associated with the emitter. Processing circuitry in communication with the radio is configured to determine at least one location parameter associated with the emitter based on the first radio signal and the location information of the at least one data packet of the second radio signal.

Abstract: Methods, systems, and apparatus for performing speech recognition. In some implementations, acoustic data representing an utterance is obtained. The acoustic data corresponds to time steps in a series of time steps. One or more computers process scores indicative of the acoustic data using a recurrent neural network to generate a sequence of outputs. The sequence of outputs indicates a likely output label from among a predetermined set of output labels. The predetermined set of output labels includes output labels that respectively correspond to different linguistic units and to a placeholder label that does not represent a classification of acoustic data. The recurrent neural network is configured to use an output label indicated for a previous time step to determine an output label for the current time step. The generated sequence of outputs is processed to generate a transcription of the utterance, and the transcription of the utterance is provided.

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: A method for estimating an embedding capacity includes receiving, at a deterministic reference encoder, a reference audio signal, and determining a reference embedding corresponding to the reference audio signal, the reference embedding having a corresponding embedding dimensionality. The method also includes measuring a first reconstruction loss as a function of the corresponding embedding dimensionality of the reference embedding and obtaining a variational embedding from a variational posterior. The variational embedding has a corresponding embedding dimensionality and a specified capacity. The method also includes measuring a second reconstruction loss as a function of the corresponding embedding dimensionality of the variational embedding and estimating a capacity of the reference embedding by comparing the first measured reconstruction loss for the reference embedding relative to the second measured reconstruction loss for the variational embedding having the specified capacity.

Abstract: Systems and methods for text-to-speech with novel speakers can obtain text data and output audio data. The input text data may be input along with one or more speaker preferences. The speaker preferences can include speaker characteristics. The speaker preferences can be processed by a machine-learned model conditioned on a learned prior distribution to determine a speaker embedding. The speaker embedding can then be processed with the text data to generate an output that includes audio data descriptive of the text data spoken by a novel speaker.

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.