Abstract: A system is described herein. The system includes at least one hardware processor that is configured to identify a pre-determined acoustic barrier filter, wherein the acoustic barrier filter coincides with the physical acoustic barrier and receive an audio signal within a time window at the first microphone and the second microphone. The hardware processor is also configured to calculate a first measure of variability, a second measure of variability, a third measure of variability, and a fourth measure of variability. The hardware processor further concatenates the first measure of variability, the second measure of variability, the third measure of variability, and the fourth measure of variability to form a feature vector, and inputs the feature vector into a location classifier to obtain an audio source location.

Abstract: An apparatus for autonomous vehicles includes a perception pipeline having independent classification processes operating in parallel to respectively identify objects based on sensor data flows from multiple ones of a plurality of sensors. The apparatus also includes a sensor monitoring stage to operate in parallel with the perception pipeline and to use the sensor data flows to estimate and track a confidence level of each of the plurality of different sensors, and nullify a deficient sensor when the confidence level associated with the deficient sensor fails to meet a confidence threshold.

Abstract: An apparatus to facilitate fractional convolutional kernels is disclosed. The apparatus includes one or more processors comprising a convolution circuit of a neural network, the convolution circuit to initialize a set of parameters of a fractional convolutional kernel, the set of parameters comprising at least a fractional derivative parameter that is initialized with a fractional value, and apply the fractional convolutional kernel to input data to convolve the input data to obtain output data.

Abstract: An apparatus for autonomous vehicles includes a perception pipeline having independent classification processes operating in parallel to respectively identify objects belonging to a specific object type based on sensor data flows from multiple ones of a plurality of different types of sensors. The apparatus also includes a sensor monitoring stage to operate in parallel with the perception pipeline and to use the sensor data flows to estimate and track a confidence level of each of the plurality of different types of sensors, and nullify a deficient sensor when the confidence level associated with the deficient sensor fails to meet a confidence threshold.

Abstract: In one embodiment, an apparatus comprises a processor to: identify a workload comprising a plurality of tasks; generate a workload graph based on the workload, wherein the workload graph comprises information associated with the plurality of tasks; identify a device connectivity graph, wherein the device connectivity graph comprises device connectivity information associated with a plurality of processing devices; identify a privacy policy associated with the workload; identify privacy level information associated with the plurality of processing devices; identify a privacy constraint based on the privacy policy and the privacy level information; and determine a workload schedule, wherein the workload schedule comprises a mapping of the workload onto the plurality of processing devices, and wherein the workload schedule is determined based on the privacy constraint, the workload graph, and the device connectivity graph.

Abstract: Technologies for using a shifted neural network include a compute device to determine a shift-based activation function of the shifted neural network. The shift-based activation function is a piecewise linear approximation of a transcendental activation function and is defined by a plurality of line segments such that a corresponding slope of each line segment is a power of two. The compute device further trains the shifted neural network based on shift-based input weights and the determined shift-based activation function.

Abstract: A system is described herein. The system includes at least one hardware processor that is configured to identify a pre-determined acoustic barrier filter, wherein the acoustic barrier filter coincides with the physical acoustic barrier and receive an audio signal within a time window at the first microphone and the second microphone. The hardware processor is also configured to calculate a first measure of variability, a second measure of variability, a third measure of variability, and a fourth measure of variability. The hardware processor further concatenates the first measure of variability, the second measure of variability, the third measure of variability, and the fourth measure of variability to form a feature vector, and inputs the feature vector into a location classifier to obtain an audio source location.

Abstract: An example apparatus for detecting speech includes an image receiver to receive depth information corresponding to a face. The apparatus also includes a landmark detector to detect the face comprising lips and track a plurality of descriptor points comprising lip descriptor points located around the lips. The apparatus further includes a descriptor computer to calculate a plurality of descriptor features based on the tracked descriptor points. The apparatus includes a pattern generator to generate a visual pattern of the descriptor features over time. The apparatus also further includes a speech recognition engine to detect speech based on the generated visual pattern.

Abstract: The apparatus includes an apparatus for harvesting energy. The apparatus includes a textile having an insulating substrate, a direct current (DC) power bus structure disposed in the insulating substrate, and multiple transducers. The DC power bus includes a positive conductor and a ground conductor. The transducers are secured to the insulating substrate and in electrical contact with the positive conductor and the ground conductor. Additionally, the DC bus remains conductively coupled to the transducers remaining in the textile after the textile is cut.

Abstract: The apparatus includes an apparatus for harvesting energy. The apparatus includes a textile having an insulating substrate, a direct current (DC) power bus structure disposed in the insulating substrate, and multiple transducers. The DC power bus includes a positive conductor and a ground conductor. The transducers are secured to the insulating substrate and in electrical contact with the positive conductor and the ground conductor. Additionally, the DC bus remains conductively coupled to the transducers remaining in the textile after the textile is cut.

Abstract: The present disclosure pertains to a system for voice capture via nasal vibration sensing. A system worn by a user may be able to sense vibrations through the nose of the user when the user speaks, generate an electronic signal based on the sensed vibration and generate voice data based on the electronic signal. In this manner, the system may capture a user's voice while also screening out external noise (e.g., based on the sound dampening properties of the human skull). An example system may include a wearable frame (e.g., eyeglass frame) on which is mounted at least one sensor and a device. The at least one sensor may sense vibration in the nose of a user and may generate the electronic signal based on the vibration. The device may receive the electronic signal from the at least one sensor and may generate voice data based on the electronic signal.

Abstract: An example apparatus for detecting speech includes an image receiver to receive depth information corresponding to a face. The apparatus also includes a landmark detector to detect the face comprising lips and track a plurality of descriptor points comprising lip descriptor points located around the lips. The apparatus further includes a descriptor computer to calculate a plurality of descriptor features based on the tracked descriptor points. The apparatus includes a pattern generator to generate a visual pattern of the descriptor features over time. The apparatus also further includes a speech recognition engine to detect speech based on the generated visual pattern.

Abstract: A voice capture system worn by a user senses vibrations through the nose of the user when the user speaks. The voice capture system generates an electronic signal based on the sensed vibration and generate voice data based on the electronic signal. In this manner, the system may capture a user's voice while also screening out external noise (e.g., based on the sound dampening properties of the human skull). In one example, a voice capture system may include a wearable frame (e.g., eyeglass frame) on which is mounted at least one sensor and a device. The at least one sensor senses vibration in the nose of a user and generates the electronic signal based on the vibration. The device receives the electronic signal from the at least one sensor and generates voice data based on the electronic signal.

Abstract: A system for sound capture and generation via nasal vibration is described. In embodiments the system includes eyeglasses that include at least a frame that is wearable by a user. Sensing circuitry is mounted to the frame, and a device is incorporated into the frame. The sensing circuitry includes at least one sensor, wherein the sensor can passively sense voice vibration induced in the user's nose, and/or which may actively induce audio vibration in the user's nose based on audio data.

Abstract: Technologies for using a shifted neural network include a compute device to determine a shift-based activation function of the shifted neural network. The shift-based activation function is a piecewise linear approximation of a transcendental activation function and is defined by a plurality of line segments such that a corresponding slope of each line segment is a power of two. The compute device further trains the shifted neural network based on shift-based input weights and the determined shift-based activation function.

Abstract: This disclosure pertains to a system for sound capture and generation via nasal vibration. An example system to capture and generate sound may comprise at least a frame wearable by a user, sensing circuitry mounted to the frame and a device also mounted to the frame. The sensing circuitry may sense voice vibration induced in the user's nose by the user's voice, generate an electronic signal based on the sensed voice vibration and induce audio vibration in the nose based on audio data. The device may be to at least control the operation of the sensing circuitry. The sensing circuitry may comprise at least one piezoelectric diaphragm to generate the electronic signal and induce the audio vibration. In at least one example implementation, the frame may be for eyeglasses and may comprise at least one nosepiece structure for contacting the nose, the at least one structure including the sensing circuitry.

Abstract: A voice capture system worn by a user senses vibrations through the nose of the user when the user speaks. The voice capture system generates an electronic signal based on the sensed vibration and generate voice data based on the electronic signal. In this manner, the system may capture a user's voice while also screening out external noise (e.g., based on the sound dampening properties of the human skull). In one example, a voice capture system may include a wearable frame (e.g., eyeglass frame) on which is mounted at least one sensor and a device. The at least one sensor senses vibration in the nose of a user and generates the electronic signal based on the vibration. The device receives the electronic signal from the at least one sensor and generates voice data based on the electronic signal.