Abstract: A forklift autonomous operation system is disclosed. The forklift autonomous operation system includes a forklift having a load handling system, the load handling system including a mast and a plurality of forks and a camera, coupled to the load handling system, for obtaining visual input data of an environment. Further, the system includes a plurality of sensors coupled to the forklift for obtaining sensor data and a control system configured to process the visual input data and the sensor data.

Abstract: A computing system is configured to obtain a plurality of media files that each includes speech of one or more speakers. The computing system is further configured to process the plurality of media files to generate indexed data, wherein the indexed data includes a corresponding embedding for each speaker of the one or more speakers identified in the media file and a corresponding one or more keywords identified in the speech in the media file. The computing system is further configured to receive an indication at least one of a selection of a particular speaker from the one or more speakers or a selection of a particular keyword from a plurality of keywords. The computing system is further configured to generate one or more correlations based on the indexed data. The computing system is further configured to output an alert regarding the one or more correlations.

Abstract: In general, the disclosure describes a computing system to automatically identify and classify audio input, including non-speech audio signals. The computing system may also add new classes based on only a limited number of examples of the new classes, to identify classes of sounds for which the system had not been trained.

Abstract: In general, the disclosure describes techniques for detecting synthetic speech of a speaker. In an example, a machine learning system may be configured to generate, using a deep learning model trained to distinguish between synthetic speech and authentic speech, reference embeddings for the speaker that characterize a first set of acoustic features and a first set of phonetic features associated with the speaker. The machine learning system may further be configured to generate, using the deep learning model, a test embedding for an audio clip that characterizes a second set of acoustic features and a second set of phonetic features associated with the audio clip. The machine learning system may further be configured to compute a score based on the test embedding and the reference embeddings. The machine learning system may further be configured to output, based on the score, an indication of whether the audio clip includes synthetic speech.

Abstract: Methods, systems, and media for presenting annotations across multiple videos are provided. In some implementations, methods for presenting annotations across multiple videos are provided, the methods comprising: receiving a selection of an annotation from a user; identifying, by a hardware processor, a plurality of videos associated with the user account to be associated with the annotation; and presenting the annotation upon rendering one of the plurality of videos.

Abstract: Systems, methods, and computer program products for programming a controller to control a process. A training module is trained to provide predicted responses of the process to control signals using a generic simulator to model the process. Once the training module has been trained, the controller is programmed to control the process by coupling the controller to the training module and using the training module to generate predicted responses of the process to the reception of control signals from the controller. As part of the programming process, one or more parameters of a control program resident on the controller may be adjusted to reduce errors between the responses predicted by the training module and the desired responses intended to be caused by the control signals.

Abstract: Methods, systems, and media for presenting annotations across multiple videos are provided. In some implementations, methods for presenting annotations across multiple videos are provided, the methods comprising: receiving a selection of an annotation from a user; identifying, by a hardware processor, a plurality of videos associated with the user account to be associated with the annotation; and presenting the annotation upon rendering one of the plurality of videos.

Abstract: A sound making device and system for generating white noise is disclosed. The sound making device generally includes a substantially dome-shaped device that includes an outer acoustic shell and an inner acoustic shell mounted atop a base tray. Both the outer acoustic shell and inner acoustic shell have slots (or openings) that overlap to create apertures through which rushing air can pass. The amount of slot overlap is variable. Further, a variable speed fan is arranged inside the outer acoustic shell and inner acoustic shell for forcing airflow out of the apertures. The sound making device also includes a controller and user interface for adjusting the speed of the fan. Further, a sound making system is provided that includes the sound making device in combination with a mobile app.

Abstract: Methods, systems, and media for presenting annotations across multiple videos are provided. In some implementations, methods for presenting annotations across multiple videos are provided, the methods comprising: receiving a selection of an annotation from a user; identifying, by a hardware processor, a plurality of videos associated with the user account to be associated with the annotation; and presenting the annotation upon rendering one of the plurality of videos.

Abstract: A spoken command analyzer computing system includes technologies configured to analyze information extracted from a speech sample and, using a joint speaker and phonetic content model, both determine whether the analyzed speech includes certain content (e.g., a command) and to identify the identity of the human speaker of the speech. In response to determining that the identity matches the authorized user's identity and determining that the analyzed speech includes the modeled content (e.g., command), an action corresponding to the verified content (e.g., command) is performed by an associated device.

Abstract: Systems, methods, and computer program products for programming a controller to control a process. A training module is trained to provide predicted responses of the process to control signals using a generic simulator to model the process. Once the training module has been trained, the controller is programmed to control the process by coupling the controller to the training module and using the training module to generate predicted responses of the process to the reception of control signals from the controller. As part of the programming process, one or more parameters of a control program resident on the controller may be adjusted to reduce errors between the responses predicted by the training module and the desired responses intended to be caused by the control signals.

Abstract: A sound making device and system for generating white noise is disclosed. The sound making device generally includes a substantially dome-shaped device that includes an outer acoustic shell and an inner acoustic shell mounted atop a base tray. Both the outer acoustic shell and inner acoustic shell have slots (or openings) that overlap to create apertures through which rushing air can pass. The amount of slot overlap is variable. Further, a variable speed fan is arranged inside the outer acoustic shell and inner acoustic shell for forcing airflow out of the apertures. The sound making device also includes a controller and user interface for adjusting the speed of the fan. Further, a sound making system is provided that includes the sound making device in combination with a mobile app.

Abstract: Methods, systems, and media for presenting annotations across multiple videos are provided. In some implementations, methods for presenting annotations across multiple videos are provided, the methods comprising: receiving a selection of an annotation from a user; identifying, by a hardware processor, a plurality of videos associated with the user account to be associated with the annotation; and presenting the annotation upon rendering one of the plurality of videos.

Abstract: An audio file analyzer computing system includes technologies to, among other things, localize audio events of interest (such as speakers of interest) within an audio file that includes multiple different classes (e.g., different speakers) of audio. The illustrative audio file analyzer computing system uses a seed segment to perform a semi-supervised diarization of the audio file. The seed segment is pre-selected, such as by a human person using an interactive graphical user interface.

Abstract: Methods, systems, and media for presenting annotations across multiple videos are provided. In some implementations, methods for presenting annotations across multiple videos are provided, the methods comprising: presenting a user interface to a user, wherein the user interface prompts the user to select an annotation to be associated with a plurality of videos associated with a user account; receiving a selection of an annotation to associate with the plurality of videos associated with the user account; transmitting data to a server identifying the selected annotation; and causing the server to associate the annotation with the plurality of videos. In some implementations, the methods further comprise: communicating with the server to set placement and timing information of the annotation; and causing the annotation to be presented upon rendering one of the plurality of videos based on the placement information of the annotation and the timing information of the annotation.

Abstract: Methods, systems, and media for presenting annotations across multiple videos are provided. In some implementations, methods for presenting annotations across multiple videos are provided, the methods comprising: receiving a selection of an annotation from a user; identifying, by a hardware processor, a plurality of videos associated with the user account to be associated with the annotation; and presenting the annotation upon rendering one of the plurality of videos.

Abstract: An audio file analyzer computing system includes technologies to, among other things, localize audio events of interest (such as speakers of interest) within an audio file that includes multiple different classes (e.g., different speakers) of audio. The illustrative audio file analyzer computing system uses a seed segment to perform a semi-supervised diarization of the audio file. The seed segment is pre-selected, such as by a human person using an interactive graphical user interface.

Abstract: A spoken command analyzer computing system includes technologies configured to analyze information extracted from a speech sample and, using a joint speaker and phonetic content model, both determine whether the analyzed speech includes certain content (e.g., a command) and to identify the identity of the human speaker of the speech. In response to determining that the identity matches the authorized user's identity and determining that the analyzed speech includes the modeled content (e.g., command), an action corresponding to the verified content (e.g., command) is performed by an associated device.

Abstract: A compressor blade having an airfoil that comprises oppositely-disposed convex and concave surfaces, oppositely-disposed leading and trailing edges defining therebetween a chord length of the airfoil, a forward-most nose of the airfoil located at the leading edge and having a profile, a blade tip, and an erosion-resistant coating. The coating is present on the concave surface near the trailing edge, optionally present on the nose, optionally present on the convex surface, wherein the convex surface is free of the erosion resistant coating within at least 20% of the chord length from the nose. The thickness of the coating on the concave surface, the convex surface, and the nose is such that, if the gas turbine engine is exposed to an erosive environment, deterioration of the concave surface, the convex surface and the leading edge does not form a pronounced cusp at an intersection of the convex surface and leading edge.

Abstract: A process for depositing coatings, and particularly erosion-resistant coatings suitable for protecting surfaces subjected to collisions with particles, such as a compressor blade of a gas turbine engine. The blade has an airfoil comprising oppositely-disposed convex and concave surfaces, oppositely-disposed leading and trailing edges defining therebetween a chord length of the airfoil, and a blade tip. An erosion-resistant coating is present on at least the concave surface, but not on the convex surface within at least 20% of the chord length from the leading edge.