Abstract: A method, computer system, and a computer program product for speech synthesis is provided. The present invention may include generating one or more final voiceprints. The present invention may include generating one or more voice clones based on the one or more final voiceprints. The present invention may include classifying the one or more voice clones into a grouping using a language model, wherein the language model is trained using manually classified uncloned voice samples. The present invention may include identifying a cluster within the grouping, wherein the cluster is identified by determining a difference between corresponding vectors of the one or more voice clones below a similarity threshold. The present invention may include generating a new archetypal voice by blending the one or more voice clones of the cluster where the difference between the corresponding vectors is below the similarity threshold.

Abstract: A monitoring system that is configured to monitor a property is disclosed. In one aspect, the monitoring system includes one or more sensors that are located throughout the property and that are configured to generate sensor data. The monitoring system further includes a drone that is configured to move throughout the property and generate additional sensor data. The monitoring system further includes a drone dock that is configured to receive the drone, wherein the drone is configured to continue generating the additional sensor data while the drone dock is receiving the drone. The monitoring system further includes a monitor control unit that is configured to receive the sensor data and the additional sensor data, analyze the sensor data and the additional sensor data, determine a status of the property, and provide, for output, data indicating the status of the property.

Abstract: Certain aspects of the present disclosure provide techniques for determining a pose of a three-dimensional deformable object. Embodiments include providing one or more inputs to a machine learning model based on a computer-generated three-dimensional deformable object that has a known pose. Embodiments include determining, based on one or more outputs from the machine learning model in response to the one or more inputs, a two-dimensional signature of the computer-generated three-dimensional deformable object. Embodiments include associating the two-dimensional signature with the known pose of the computer-generated three-dimensional deformable object. Embodiments include determining a respective pose of an actual three-dimensional deformable object based on an image of the actual three-dimensional deformable object and the associating.

Abstract: Certain aspects of the present disclosure provide techniques for generating three-dimensionally coherent training data for image-detection machine learning models. Embodiments include receiving a first image of an object from a first perspective and a second image of the object from a second perspective. Embodiments include receiving user input identifying a location in the first image corresponding to a point on the object. Embodiments include displaying a range of possible locations in the second image corresponding to the point on the object based on the location in the first image. Embodiments include generating training data for a machine learning model based on updated user input associated with the range of possible locations.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a storage device, for enabling drone activity in a property monitored by a property monitoring system without triggering a false alarm. In one aspect, the method includes actions of obtaining a location of the drone, identifying a first sensor installed at the property that is within a predetermined distance of the drone, detecting first sensor data generated by the first sensor that is within a predetermined distance of the drone, wherein the first sensor data includes data that is indicative of an event, determining whether a second sensor that is mounted to the drone is generating second sensor data that corroborates the event indicated by the first sensor data, and disregarding, by the property monitoring system, the first sensor data in determining whether to trigger an alarm.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for reduction of false detections using an ultrasound emitter. One of the methods includes obtaining sensor data captured by a sensor of a property monitoring system; detecting an object represented by the sensor data; determining that the object represented by the sensor data is likely a non-object of interest and being detected as an object of interest; and in response to determining that the object represented by the sensor data is likely the non-object of interest and being detected as the object of interest, sending, to an ultrasound emitter, a command to cause the ultrasound emitter to emit ultrasonic sound around the sensor.

Abstract: A method includes generating, by a camera of a monitoring system that is configured to monitor a property, first image data at a first time, analyzing the first image data, determining that the first image data includes a first object that likely corresponds to a person, generating, second image data at a second, later time, analyzing the second image data, determining that the second image includes a second object that likely corresponds to a person, comparing the first object that likely corresponds to a person to the second object that likely corresponds to a person, based on comparing the first object that likely corresponds to a person to the second object that likely corresponds to a person and based on a difference between the first time and the second, later time, determining that a person is likely moving towards the camera, and performing a monitoring system action.

Abstract: A robotic rover that is configured to monitor a property is disclosed. In one aspect, the rover includes a sensor that is configured to generate sensor data. The rover further includes a processor that is configured to analyze the sensor data. The processor is further configured to determine that an unwanted animal is located at the property. The processor is further configured to generate an instruction to navigate the drone towards the unwanted animal. The rover further includes a motor that is configured to navigate the drone towards the unwanted animal in response to the instruction to navigate the drone towards the unwanted animal. The processor is further configured to generate an instruction to activate a device that is configured to encourage the unwanted animal to leave the property.

Abstract: Disclosed herein are system, method, and computer program product embodiments for enabling access to a firmware-locked function of a secure device. A secure device may be production hardware that has locked certain functions not available for public use. In an embodiment, the secure device may receive a request to access a firmware-locked function. The request may include an authorization token that includes an identifier specific to the particular secure device and a permission identifier identifying the firmware-locked function. Based on the receipt of the authorization token, the secure device may retrieve authorization data from the firmware memory of the secure device to determine whether the provided authorization token matches the firmware authorization data. Using the authorization token, the secure device may determine whether to grant access to the firmware-locked function.

Abstract: One example relates to a device that includes an activity monitor. The activity monitor includes a bus interface having inputs coupled to receive signals from a bus and having outputs coupled to provide signals to an other device. The activity monitor monitors the bus for a message directed to a predefined address that is associated with the other device in response to detecting that the other device is a low power sleep mode, the activity monitor outputs the predefined address to the other device to enable the other device to capture the predefined address, via the bus interface, in response to the monitored address matching the predefined address that is associated with the other device.

Abstract: Disclosed herein are various embodiments of an electronic lockbox for storing keys and key cards. The lockbox may include calendar-based entry rules so that the key codes can differ based on the date and time. The lockbox may be controllable by a mobile device such as a mobile phone, tablet, or other mobile device. The lockbox may include functionality so that key codes for the lockbox can be transmitted via websites such as Airbnb, HomeAway, or other rental websites, so that guests automatically receive their unique lockbox codes when they confirm their reservations. The lockbox may also receive a Bluetooth low energy virtual key which may be transmitted directly from a user's mobile phone to the lockbox, so that no manual code entry through the keypad is required by a user. The lockbox may be capable of accepting multiple passcodes enabled at different times and dates, and for different purposes.

Abstract: A computer implemented method including receiving, by a monitoring system that is configured to monitor a property and from an electronic pool device that is configured to monitor a swimming pool at the property, sensor data, analyzing, by the monitoring system, the sensor data, based on analyzing the sensor data, generating, by the monitoring system, an instruction to activate a camera of the electronic pool device, providing, by the monitoring system to the electronic pool device, the instruction to activate the camera, receiving, by the monitoring system from the electronic pool device, image data, analyzing, by the monitoring system, the image data, based on analyzing the image data, identifying a monitoring system action to perform, and performing the monitoring system action.

Abstract: A power circuit includes a switch circuit, an auxiliary load circuit coupled to an output terminal, a switching control circuit to operate the switch circuit responsive to an error signal, a regulator circuit having a sense resistor, a comparator to provide the error signal, and a DAC to control a sense current of the sense resistor. A DAC control circuit provides a DAC input signal having a controlled ramp rate responsive to a decreasing setpoint signal, a load control circuit selectively enables the auxiliary load circuit responsive to the decreasing setpoint signal and responsive to the error signal to control the power circuit slew rate.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a storage device, for using a drone to test a sensor. In one aspect, the method includes actions of detecting a message (i) broadcast by the drone and (ii) indicating that the drone is going to administer a test of a sensor, determining, by the monitoring system and based on the message and (i) sensor data generated by the sensor in response to the administration of the test, by the drone, within a predetermined period of time of the message or (ii) a lack of sensor data generated by the sensor in response to the administration of the test, by the drone, within a predetermined period of time of the message, whether the sensor is functioning properly, and in response to a determination that the sensor is not functioning properly, storing data indicating that the sensor is not functioning properly.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for reduction of false detections using an ultrasound emitter. One of the methods includes obtaining sensor data captured by a sensor of a property monitoring system; detecting an object represented by the sensor data; determining that the object represented by the sensor data is likely a non-object of interest and being detected as an object of interest; and in response to determining that the object represented by the sensor data is likely the non-object of interest and being detected as the object of interest, sending, to an ultrasound emitter, a command to cause the ultrasound emitter to emit ultrasonic sound around the sensor.

Abstract: One example relates to a device that includes an activity monitor. The activity monitor includes a bus interface having inputs coupled to receive signals from a bus and having outputs coupled to provide signals to an other device. The activity monitor monitors the bus for a message directed to a predefined address that is associated with the other device in response to detecting that the other device is a low power sleep mode, the activity monitor outputs the predefined address to the other device to enable the other device to capture the predefined address, via the bus interface, in response to the monitored address matching the predefined address that is associated with the other device.

Abstract: A monitoring system that is configured to monitor a property is disclosed. In one aspect, the monitoring system includes one or more sensors that are located throughout the property and that are configured to generate sensor data. The monitoring system further includes a drone that is configured to move throughout the property and generate additional sensor data. The monitoring system further includes a drone dock that is configured to receive the drone, wherein the drone is configured to continue generating the additional sensor data while the drone dock is receiving the drone. The monitoring system further includes a monitor control unit that is configured to receive the sensor data and the additional sensor data, analyze the sensor data and the additional sensor data, determine a status of the property, and provide, for output, data indicating the status of the property.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a storage device, for enabling drone activity in a property monitored by a property monitoring system without triggering a false alarm. In one aspect, the method includes actions of obtaining a location of the drone, identifying a first sensor installed at the property that is within a predetermined distance of the drone, detecting first sensor data generated by the first sensor that is within a predetermined distance of the drone, wherein the first sensor data includes data that is indicative of an event, determining whether a second sensor that is mounted to the drone is generating second sensor data that corroborates the event indicated by the first sensor data, and disregarding, by the property monitoring system, the first sensor data in determining whether to trigger an alarm.

Abstract: An inductor has first and second terminals. A first switch is coupled between the first terminal and a voltage supply terminal. A second switch is coupled between the first terminal and a negative output supply terminal. A third switch is coupled between the second terminal and a positive output supply terminal. A fourth switch is coupled between the second terminal and a ground terminal. A controller is coupled to the first, second, third and fourth switches. The controller is configured to provide: an inductor charge mode; a positive boost mode; a negative boost mode; a first rest state in which the controller closes the first switch and opens the second, third and fourth switches; and a second rest state in which the controller closes the fourth switch and opens the first, second and third switches.

Abstract: A computer-implemented method including receiving, by a monitoring system that is configured to monitor a property and from a touchless doorbell device that includes a camera and a motion sensor, an image, analyzing, by the monitoring system, the image, based on analyzing the image, determining that the image likely includes a representation of a person, and based on determining that the image likely includes a representation of a person, performing a monitoring system action.