Abstract: A method to detect a vocal command, the method including: analyzing audio data received from a transducer configured to convert audio into an electric signal and analyzing the data using a first neural network. The method also includes detecting a keyword from the audio data using the first neural network on the edge device, the first neural network being trained to recognize the keyword. The method further includes activating a second neural network after the keyword is identified by the first neural network and analyzing the audio data using the second neural network, the second neural network being trained to recognize a set of vocal commands. The method to detect a vocal command may also include detecting the vocal command word using the second neural network.

Abstract: A method of processing an electrical signal transduced from a voice signal is disclosed. A classification model is applied to the electrical signal to produce a classification indicator. The classification model has been trained using an augmented training dataset. The electrical signal is classified as either one of a first class and a second class in a binary classification. The classifying being performed is a function of the classification indicator. A trigger signal is provided to a user circuit as a result of the electrical signal being classified in the first class of the binary classification.

Abstract: Voice command recognition and natural language recognition are carried out using an accelerometer that senses signals from the vibrations of one or more bones of a user and receives no audio input. Since word recognition is made possible using solely the signal from the accelerometer from a person's bone conduction as they speak, an acoustic microphone is not needed and thus not used to collect data for word recognition. According to one embodiment, a housing contains an accelerometer and a processor, both within the same housing. The accelerometer is preferably a MEMS accelerometer which is capable of sensing the vibrations that are present in the bone of a user as the user is speaking words. A machine learning algorithm is applied to the collected data to correctly recognize words spoken by a person with significant difficulties in creating audible language.

Abstract: A method to detect a vocal command, the method including: analyzing audio data received from a transducer configured to convert audio into an electric signal and analyzing the data using a first neural network. The method also includes detecting a keyword from the audio data using the first neural network on the edge device, the first neural network being trained to recognize the keyword. The method further includes activating a second neural network after the keyword is identified by the first neural network and analyzing the audio data using the second neural network, the second neural network being trained to recognize a set of vocal commands. The method to detect a vocal command may also include detecting the vocal command word using the second neural network.

Abstract: In accordance with embodiments, methods and systems for a trigger to the KWS are provided. The computing device converts an audio signal into a plurality of audio frames. The computing device generates a Mel Frequency Cepstral Coefficients (MFCC) matrix. The MFCC matrix includes N columns. Each column of the N columns comprises coefficients associated with audio features corresponding to a different audio frame of the plurality of audio frames. The computing device determines that a trigger condition is satisfied based on an MFCC_0 buffer. The MFCC_0 buffer comprises a first row of the MFCC matrix. The computing device then provides the MFCC matrix to a neural network for the neural network to use the MFCC matrix to make keyword inference based on the determining that the trigger condition is satisfied.

Abstract: A capacitive touch screen of e.g., a mobile communications device such as a smart phone or tablet is operated by producing a capacitance map of capacitance values for the screen, wherein the capacitance values are indicative of locations of the screen exposed to touch by a user, and by identifying locations of the screen exposed to touch by a user by comparing the capacitance values against settings of sensing thresholds. Descriptor processing is applied to the capacitance map to extract a set of descriptors indicative of said screen being in one of a plurality of different operating conditions. A set of rules is applied to these descriptors to identify one of a plurality of different operating conditions, and selecting the setting of sensing thresholds as a function of the operating condition thus identified.

Abstract: In accordance with embodiments, methods and systems for a trigger to the KWS are provided. The computing device converts an audio signal into a plurality of audio frames. The computing device generates a Mel Frequency Cepstral Coefficients (MFCC) matrix. The MFCC matrix includes N columns. Each column of the N columns comprises coefficients associated with audio features corresponding to a different audio frame of the plurality of audio frames. The computing device determines that a trigger condition is satisfied based on an MFCC_0 buffer. The MFCC_0 buffer comprises a first row of the MFCC matrix. The computing device then provides the MFCC matrix to a neural network for the neural network to use the MFCC matrix to make keyword inference based on the determining that the trigger condition is satisfied.

Abstract: A method of processing an electrical signal transduced from a voice signal is disclosed. A classification model is applied to the electrical signal to produce a classification indicator. The classification model has been trained using an augmented training dataset. The electrical signal is classified as either one of a first class and a second class in a binary classification. The classifying being performed is a function of the classification indicator. A trigger signal is provided to a user circuit as a result of the electrical signal being classified in the first class of the binary classification.

Abstract: Hand gestures, such as hand or finger hovering, in the proximity space of a sensing panel are detected from X-node and Y-node sensing signals indicative of the presence of a hand feature at corresponding row and column locations of a sensing panel. Hovering is detected by detecting the locations of maxima for a plurality of frames over a time window for sets of X-node and Y-node sensing signals by recognizing a hovering gesture if the locations of the maxima detected vary over the plurality of frames for one of the sets of sensing signals and not for the other of set. Finger shapes are distinguished over “ghosts” generated by palm or fist features by transforming the node-intensity representation for the sensing signals into a node-distance representation based on distances of detection intensities for a number of nodes under a peak for a mean point between valleys adjacent to the peak.

Abstract: Method for generating a lane departure warning in a vehicle, comprising acquiring a plurality of frames of a digital image of a road on which the vehicle is running, the digital image of a road including the image of a lane within which the vehicle is running and of marking lines of the lane, for each of the acquired frames, extracting edge points of the frame, analyzing the edge points to evaluate a lane departure status, the evaluation including performing a lane departure verification procedure including identifying in the frame points representative of the position of the lane marking lines, generating a lane departure alert if a lane departure status is detected by the lane departure verification procedure.

Abstract: A sequence of images is processed to generate optical flow data including a list of motion vectors. The motion vectors are grouped based on orientation into a first set of moving away motion vectors and a second set of moving towards motion vectors. A vanishing point is determined as a function of the first set of motion vectors and a center position of the images is determined. Pan and tilt information is computed from the distance difference between the vanishing point and the center position. Approaching objects are identified from the second set as a function of position, length and orientation, thereby identifying overtaking vehicles. Distances to the approaching objects are determined from object position, camera focal length, and pan and tilt information. A warning signal is issued as a function of the distances.

Abstract: A sequence of images obtained by a camera mounted on a vehicle is processed in order to generate Optical Flow data including a list of Motion Vectors being associated with respective features in the sequence of images. The Optical Flow data is analyzed to calculate a Vanishing Point by calculating the mean point of all intersections of straight lines passing through motion vectors lying in a road. An Horizontal Filter subset is determined taking into account the Vanishing Point and a Bound Box list from a previous frame in order to filter from the Optical Flow the horizontal motion vectors. The subset of Optical Flow is clustered to generate the Bound Box list retrieving the moving objects in a scene. The Bound Box list is sent to an Alert Generation device and an output video shows the input scene where the detected moving objects are surrounded by a Bounding Box.

Abstract: Method for generating a lane departure warning in a vehicle, comprising acquiring a plurality of frames of a digital image of a road on which the vehicle is running, the digital image of a road including the image of a lane within which the vehicle is running and of marking lines of the lane, for each of the acquired frames, extracting edge points of the frame, analyzing the edge points to evaluate a lane departure status, the evaluation including performing a lane departure verification procedure including identifying in the frame points representative of the position of the lane marking lines, generating a lane departure alert if a lane departure status is detected by the lane departure verification procedure.

Abstract: Hand gestures, such as hand or finger hovering, in the proximity space of a sensing panel are detected from X-node and Y-node sensing signals indicative of the presence of a hand feature at corresponding row and column locations of a sensing panel. Hovering is detected by detecting the locations of maxima for a plurality of frames over a time window for sets of X-node and Y-node sensing signals by recognizing a hovering gesture if the locations of the maxima detected vary over the plurality of frames for one of the sets of sensing signals and not for the other of set. Finger shapes are distinguished over “ghosts” generated by palm or fist features by transforming the node-intensity representation for the sensing signals into a node-distance representation based on distances of detection intensities for a number of nodes under a peak for a mean point between valleys adjacent to the peak.

Abstract: In an embodiment, hand gestures, such as hand or finger hovering, in the proximity space of a sensing panel are detected from X-node and Y-node sensing signals indicative of the presence of a hand feature at corresponding row locations and column locations of a sensing panel. Hovering is detected by detecting the locations of maxima for a plurality of frames over a time window for a set of X-node sensing signals and for a set of Y-node sensing signals by recognizing a hovering gesture if the locations of the maxima detected vary over the plurality of frames for one of the sets of X-node and Y-node sensing signals while remaining stationary for the other of the sets of X-node and Y-node sensing signals.

Abstract: A sequence of images is processed to generate optical flow data including a list of motion vectors. The motion vectors are grouped based on orientation into a first set of moving away motion vectors and a second set of moving towards motion vectors. A vanishing point is determined as a function of the first set of motion vectors and a center position of the images is determined. Pan and tilt information is computed from the distance difference between the vanishing point and the center position. Approaching objects are identified from the second set as a function of position, length and orientation, thereby identifying overtaking vehicles. Distances to the approaching objects are determined from object position, camera focal length, and pan and tilt information. A warning signal is issued as a function of the distances.

Abstract: A sequence of images obtained by a camera mounted on a vehicle is processed in order to generate Optical Flow data including a list of Motion Vectors being associated with respective features in the sequence of images. The Optical Flow data is analyzed to calculate a Vanishing Point by calculating the mean point of all intersections of straight lines passing through motion vectors lying in a road. An Horizontal Filter subset is determined taking into account the Vanishing Point and a Bound Box list from a previous frame in order to filter from the Optical Flow the horizontal motion vectors. The subset of Optical Flow is clustered to generate the Bound Box list retrieving the moving objects in a scene. The Bound Box list is sent to an Alert Generation device and an output video shows the input scene where the detected moving objects are surrounded by a Bounding Box.

Abstract: In an embodiment, hand gestures, such as hand or finger hovering, in the proximity space of a sensing panel are detected from X-node and Y-node sensing signals indicative of the presence of a hand feature at corresponding row locations and column locations of a sensing panel. Hovering is detected by detecting the locations of maxima for a plurality of frames over a time window for a set of X-node sensing signals and for a set of Y-node sensing signals by recognizing a hovering gesture if the locations of the maxima detected vary over the plurality of frames for one of the sets of X-node and Y-node sensing signals while remaining stationary for the other of the sets of X-node and Y-node sensing signals(Y).

Abstract: An embodiment of a method for processing finger-detection data produced by a touch screen includes: computing the area of the finger-data map and extracting the main axes from the finger-data map, computing the lengths and orientations of the main axes, determining from the main axes a major axis having a major-axis orientation, computing a geometrical center and a center of mass of the finger-data map, computing an eccentricity of the finger-data map as a function of the lengths of the main axes outputting the major-axis orientation as indicative of the finger-orientation direction in the plane of the screen, outputting the mutual position of the geometrical center and the center of mass of the finger-data map as indicative of finger-pointing direction along the finger-orientation direction in the plane of the screen, and outputting a combination of the eccentricity and the area of the finger data map as indicative of finger orientation with respect to the plane of the screen.

Abstract: Detecting the presence of a finger in proximity of a screen that generates detection signals in the horizontal direction and vertical direction includes sampling the detection signals and generating raw-data vectors X and Y. The raw data have a maximum for elements of the vector that define the position of the finger on the screen in the directions X and Y, respectively. The vectors X and Y are divided into subsets defined as “macro-areas” and cumulative values computed of each macro-area by adding together all the elements of the vector X and of the vector Y that belong to the macro-area. The maximum values are selected from among horizontal cumulative values and vertical cumulative values. A value identifying the macro-area selected on the basis of the maximum values is supplied, or no value supplied in the presence of elements of disturbance in the proximity of the screen.