Abstract: A device, a computer implemented method, and a computer program product, the device including a capture device configured for capturing an image, and a processor configured to: receiving at least a part of an image captured by the capture device, analyzing the at least part of the image to determine a looking direction of a subject at an object, based on the looking direction, determining an engagement level of a subject with the object, and outputting the engagement level or whether the engagement level is in compliance with a required engagement level, while disabling transmission of information that enables reconstruction of the at least part of the image.

Abstract: A method, device and computer program product, the method comprising: obtaining access to a classifier trained upon a multiplicity of sets of decoded coefficients; obtaining a set of block coefficients associated with at least a part of the compressed image; and applying the classifier to the set of block coefficients, to obtain a classification of the compressed image.

Abstract: A method for controlling a bias of a neural network, the method may include training the neural network by using a loss function that is responsive to classes of a classification process and sensitivity input values thereby setting the bias.

Abstract: This disclosure provides methods, devices, and systems for videoconferencing. The present implementations more specifically relate to audio signal processing techniques that can be used to identify speakers in a videoconference. In some aspects, an audio signal processor may map each speaker in a videoconference to a respective spatial direction and transform the audio signals received from each speaker using one or more transfer functions associated with the spatial direction to which the speaker is mapped. The audio signal processor may further transmit the transformed audio signals to an audio output device that emits sounds waves having a directionality associated with the transformation. For example, the audio signal processor may apply one or more head-related transfer functions to the audio signals received from a particular speaker so that the sound waves emitted by the audio output device are perceived as originating from the spatial direction to which the speaker is mapped.

Abstract: An active noise cancelling system (20) comprising: an earphone (8?) comprising: an electro-acoustic driver (11); and at least one sensing microphone (12, 13); tunable active noise cancelling circuitry (7) operative to receive a signal from the at least one sensing microphone (12, 13), the tunable active noise cancelling circuitry (7) being pre-configured in a standard tuning for a reference ear and comprising at least one noise-control filter (14, 15); and a tuning module (24) operative to configure the earphone (8?) for an individual wearer by: comparing acoustic coupling of the earphone (8?) to the individual wearer's ear with acoustic coupling to the reference ear to determine a deviation in acoustic coupling; and using the determined deviation in acoustic coupling to modify the tunable active noise cancelling circuitry (7) by a predetermined degree based on the determined deviation in acoustic coupling.

Abstract: There is provided a RF-DAC that may include (i) a first PAM that includes a first group of first power amplifiers of different amplifications, (ii) a second PAM that includes a second group of second power amplifiers of different amplifications; (iii) a load that includes an output port and a transformer; (iv) power amplifiers control units, and a transformer control unit.

Abstract: This disclosure provides methods, devices, and systems for object detection in images. The present implementations more specifically relate to cascaded object detection techniques. In some aspects, an object detection system may obtain a first image and perform a first object detection operation on the first image. The object detection system may select a region of interest (ROI) in the first image based at least in part on the object detection operation, the ROI being selected before the object detection operation is completed. The object detection system may obtain a second image coinciding with the ROI of the first image, the second image having a higher resolution than the first image. The object detection system may selectively perform an inferencing operation on the second image based on a result of the object detection operation.

Abstract: A first mobile communication device that includes a first microphone, a first speaker, and a first delay unit. The first microphone is configured to (i) receive, during a conference call, a first user first microphone signal from a first user, and (ii) output a first microphone digital signal to the first delay unit. The first user first microphone signal represents audio content outputted by the first user. The first delay unit is configured to delay, by a delay period, the first microphone digital signal to provide a delayed first user first device digital signal. The first mobile communication device is configured to output, to a mixer, the delayed first user first device digital signal. The delay period is determined based on measurements executed by at least one mobile communication device out of the first mobile communication device, a second mobile communication device and a third mobile communication device.

Abstract: A coin cell powered device includes a regulator, a transmitter and a charge pump. The transmitter is configured to transmit signals during a transmission period while receiving power from the regulator, the power originated from an external capacitor. The charge pump is configured to perform, during a charging period, a charging process for charging the external capacitor to a charged voltage that exceeds a voltage of a cell coin, wherein the charging process may include iterations of (a) charging a charge pump capacitor by the coin cell, and (b) discharging the charge pump capacitor thereby charging the external capacitor. The capacitance of the charge pump capacitor is a fraction of a capacitance of the external capacitor. The duration of the charging period exceeds a duration of the transmission period.

Abstract: A TX/RX RF switch that may include a reception path; and a transmission path that has an antenna port, a transmission input port, and transmission transistors. The transmission transistors have source-bulk connections. The reception path has an antenna port, a reception output port, and reception transistors. The reception path includes a first reception transistor that is closest to the antenna port, out of the reception transistors, and has a source-bulk connection, and at least one other reception transistor that has a bulk-to-ground connection. The reception transistors and the transmission transistors are CMOS transistors.

Abstract: A method for receiving audio broadcasts during a digital enhanced cordless telecommunication (DECT) frame, the method may include (a) opening, during a DECT frame and by a portable point (PP) that is a potential recipient of the audio broadcasts from a fixed part (FP), multiple reception windows that are spaced apart from each other and are allocated to receive signals related to audio broadcast; (b) receiving received signals, by the PP and during the multiple spaced apart reception windows; and (c) processing, by the PP, the received signals.

Abstract: A method for distributed neural network processing, the method may include detecting, by a local neural network that belongs to a local device, and based on sensed information, an occurrence of a triggering event for executing or completing a classification or detection process; sending to a remote device, a request for executing or completing the classification or detection process by a remote device that comprises a remote neural network; wherein the remote neural network has more computational resources than the local neural network; determining by the remote device whether to accept the request; and executing or completing, by the remote device, the classification or detection process when determining to accept the request; wherein the executing or completing involves utilizing the remote neural network.

Abstract: A method for in-ear detection, the method may include transmitting test signals, by a speaker of an earbud, during a test period, and while the earbud is operating at a first operational mode, wherein the test signals comprise at least one first test signal within a first frequency range, at least one second test signal within a second frequency range, and at one third test signal within a third frequency range; wherein the first frequency range, the second frequency range and the third frequency range differ from each other and are within a human auditory range; generating, by a feedback microphone of the earbud, sensed information that is indicative of audio signals sensed by the feedback microphone as a result of the transmitting of the test signals; and determining whether the earbud is located within an ear of a person, wherein the determining is based on the sensed information and a reference out of ear spectrum.

Abstract: A method for speech level adaptation, the method includes: (A) Receiving, by a voice interactive intelligent personal assistant, multiple input audio signals that includes first and second groups of input audio signals. (B) Buffering the multiple input audio signals. (C) Searching for a voice trigger in the first group of input audio signals. When finding the voice trigger then (D) Determining a linear gain factor to be applied on the second group of buffered audio signals, (E) Applying the linear gain factor on the second group of buffered audio signals to provide the output audio signals; and (F) applying a speech recognition process on the output audio signals to detect an audio command embedded in the output audio signals.

Abstract: A low noise amplifier that may include a first input port, a second input port, a first capacitor, a second capacitor, a first variable capacitor, a second variable capacitor, an inductor, a bias circuit, a tuning circuit, a first output circuit having a first output, a second output circuit having a second output; wherein the first input port is electrically coupled to a first end of the second variable capacitor, to a first end of the first capacitor, to an input of the first output circuit, and to a first port of the inductor; wherein the second input port is electrically coupled to a second end of the first variable capacitor, to a second end of the second capacitor, to an input of the second output circuit, and to a second port of the inductor; wherein a first port of the first varactor is electrically coupled to a second end of the first capacitor; wherein a second port of the second varactor is electrically coupled to a first end of the second capacitor; wherein the bias circuit is configured to supply

Abstract: A mobile communication device that comprises a microphone, a speaker, an input output port and one or more modules; wherein the microphone is configured to receive, during a conference call, (a) an acoustic signal from a first person, and (b) a cross echo from a speaker of another mobile communication device that participates in the conference call, the cross echo is generated as a result of the acoustic signal from the first person; and wherein the one or more modules are configured to cancel the cross echo based, at least in part, of the acoustic signal of the first person.

Abstract: A low noise amplifier that includes a first cascode, a second cascode, an input circuit, an output node, a first switch, and a second switch. A source of a first common gate transistor and a drain of a first common source transistor of the first cascode are coupled to a first node of the low noise amplifier. The output node is coupled to a drain of the first common gate transistor, and to a drain of a second common gate transistor of the second cascode, thereby coupling the first cascode and the second cascode to a power supply via a load. The first switch is coupled between a gate of the first common gate transistor and the power supply. The second switch is coupled between the first node and the power supply. The first switch is configured to be open and the second switch is configured to be closed when the low noise amplifier operates at a first operational node.

Abstract: A true wireless headset that may include a housing, a microphone that is mechanically coupled to the housing, wherein the microphone is configured to detect sound and output detection signals indicative of the sound, a tap detector that is configured to receive a first representation of the detection signals, and to process the first representation of the detection signals to determine whether a user tapped on the housing; and at least one additional circuitry that is configured to respond to a determination that the user tapped on the housing.

Abstract: There is provided a RF-DAC that may include (i) a first PAM that includes a first group of first power amplifiers of different amplifications, (ii) a second PAM that includes a second group of second power amplifiers of different amplifications; (iii) a load that includes an output port and a transformer; (iv) power amplifiers control units, and a transformer control unit.

Abstract: A method and a voice processor that includes (i) an input that is configured to receive of audio signals that represent audio, (ii) a wake word detection circuit, (iii) a first buffer that is configured to store at least wake word signals and prebuffer signals, and (iv) a communication module that is configured to (a) output, over an interrupt port, an interrupt request to an application processor, following a detection of the wake word signals, (b) following an acceptance of the application processor to receive content, access the first buffer and retrieve the prebuffer signals and the wake word signals; and (b) output the content, over the I2S port, to the application processor. The content includes the wake word signals, the prebuffer signals, and query or command signals.