Abstract: A method for semi-automated labeling of data for machine learning training. Data is collected via time-series sensors to form an unlabeled dataset. After receiving one or more event type labels for a subset of the dataset, thereby forming a labeled dataset, the remainder of the unlabeled dataset is automatically labeled. Potential new labels for the remainder of the unlabeled dataset are determined via cross correlation between the labeled dataset and unlabeled dataset. The potential new labels are presented as training data for a machine learning algorithm.

Abstract: Techniques enabling improved classification of touch or hover interactions of objects with a touch sensitive surface of a device are presented. A speaker of the device can emit an ultrasonic audio signal comprising a first frequency distribution. A microphone of the device can detect a reflected audio signal comprising a second frequency distribution. The audio signal can be reflected off of an object in proximity to the surface to produce the reflected audio signal. A classification component can determine movement status of the object, or classify the touch or hover interaction, in relation to the surface, based on analysis of the signals. The classification component also can classify the touch or hover interaction based on such ultrasound data and/or touch surface or other sensor data. The classification component can be trained, using machine learning, to perform classifications of touch or hover interactions of objects with the surface.

Abstract: Techniques enabling improved classification of touch or hover interactions of objects with a touch sensitive surface of a device are presented. A speaker of the device can emit an ultrasonic audio signal comprising a first frequency distribution. A microphone of the device can detect a reflected audio signal comprising a second frequency distribution. The audio signal can be reflected off of an object in proximity to the surface to produce the reflected audio signal. A classification component can determine movement status of the object, or classify the touch or hover interaction, in relation to the surface, based on analysis of the signals. The classification component also can classify the touch or hover interaction based on such ultrasound data and/or touch surface or other sensor data. The classification component can be trained, using machine learning, to perform classifications of touch or hover interactions of objects with the surface.

Abstract: Techniques enabling improved classification of touch or hover interactions of objects with a touch sensitive surface of a device are presented. A speaker of the device can emit an ultrasonic audio signal comprising a first frequency distribution. A microphone of the device can detect a reflected audio signal comprising a second frequency distribution. The audio signal can be reflected off of an object in proximity to the surface to produce the reflected audio signal. A classification component can determine movement status of the object, or classify the touch or hover interaction, in relation to the surface, based on analysis of the signals. The classification component also can classify the touch or hover interaction based on such ultrasound data and/or touch surface or other sensor data. The classification component can be trained, using machine learning, to perform classifications of touch or hover interactions of objects with the surface.

Abstract: Techniques enabling improved classification of touch or hover interactions of objects with a touch sensitive surface of a device are presented. A speaker of the device can emit an ultrasonic audio signal comprising a first frequency distribution. A microphone of the device can detect a reflected audio signal comprising a second frequency distribution. The audio signal can be reflected off of an object in proximity to the surface to produce the reflected audio signal. A classification component can determine movement status of the object, or classify the touch or hover interaction, in relation to the surface, based on analysis of the signals. The classification component also can classify the touch or hover interaction based on such ultrasound data and/or touch surface or other sensor data. The classification component can be trained, using machine learning, to perform classifications of touch or hover interactions of objects with the surface.

Abstract: Embodiments of the present invention provide methods and apparatuses for quickly and easily configuring an application user interface using a flexible generic layout file. For one embodiment, a free-form grid layout is provided that allows an application provider to create a desired number of placeholders, each of a desired size, by positioning objects at desired locations on the free-form grid. In this way the application provider configures the application user interface. For one embodiment, the placeholders are created by dragging selected objects, from a provided set of objects, onto the grid layout. For such an embodiment, a set of parameters that describe the objects on the grid layout (e.g., indicating number, size, and location) is stored to a database. At run-time, the parameters are used to dynamically generate HTML code, which when executed presents the application user interface.

Abstract: Embodiments of the present invention provide methods and apparatuses for quickly and easily configuring an application user interface using a flexible generic layout file. For one embodiment, a free-form grid layout is provided that allows an application provider to create a desired number of placeholders, each of a desired size, by positioning objects at desired locations on the free-form grid. In this way the application provider configures the application user interface. For one embodiment, the placeholders are created by dragging selected objects, from a provided set of objects, onto the grid layout. For such an embodiment, a set of parameters that describe the objects on the grid layout (e.g., indicating number, size, and location) is stored to a database. At run-time, the parameters are used to dynamically generate HTML code, which when executed presents the application user interface.

Abstract: The apparatus for the recognition of speech comprises an acoustic preprocessor, a visual preprocessor, and a speech classifier that operates the acoustic and visual preprocessed data. The acoustic preprocessor comprises a log mel spectrum analyzer that produces an equal mel bandwidth log power spectrum. The visual processor detects the motion of a set of fiducial markers on the speaker's face and extracts a set of normalized distance vectors describing lip and mouth movement. The speech classifier uses a multilevel time-delay neural network operating on the preprocessed acoustic and visual data to form an output probability distribution that indicates the probability of each candidate utterance having been spoken, based on the acoustic and visual data.