Abstract: Methods, apparatus, systems and articles of manufacture are disclosed for wear noise audio signature suppression. An example method disclosed herein includes generating an audio signature based on a media audio signal during a first time period, collecting acceleration data during the first time period, determining whether the acceleration data corresponds to wear noise having occurred during the first time period, and in response to determining the acceleration data corresponds to wear noise during the first time period, inhibiting transmission of the audio signature to a central facility.

Abstract: A voice-controlled electronic device that includes an axisymmetric device housing having a longitudinal axis bisecting opposing top and bottom surfaces and a side surface extending between the top and bottom surfaces. The device can further include a plurality of microphones disposed within the device housing and distributed radially around the longitudinal axis; a processor configured to execute computer instructions stored in a computer-readable memory for interacting with a user and processing voice commands received by the plurality of microphones and first and second transducers configured to generate sound waves within different frequency ranges.

Abstract: A method, comprising: obtaining one or more accelerometer signals derived from an accelerometer; and determining one or more parameters of wind at the accelerometer based on the one or more accelerometer signals.

Abstract: An integrated circuit includes a first input port configured to receive a supply voltage from a switched-mode power supply (SMPS), where frequency components of the supply voltage include harmonics of a reference frequency, where the reference frequency is equal to a first frequency divided by a factor; and a spurious components cancellation circuit coupled to the first input port, where the spurious components cancellation circuit is configured to: generate a first clock signal having the reference frequency; adjust an amplitude and a phase of the first clock signal to form a compensation signal; and add the compensation signal to the supply voltage to produce a modified supply voltage with reduced frequency components at one or more harmonic frequencies of the reference frequency.

Abstract: The present disclosure relates to a head unit for an infotainment system of a vehicle. The head unit is coupled to a mobile terminal device via a Bluetooth link. The head unit is configured to exchange a digital audio signal adapted for improved audio quality during the hands-free talking, via an additional wireless link, in response to hands-free talking during a phone call with the mobile terminal device coupled via the Bluetooth link based on a hands-free profile.

Abstract: The present disclosure provides for improved hot word detection in electronic devices, particularly small form factor devices such as wearables. The electronic device includes an accelerometer onboard to pick up voice in noisy conditions, and utilizes the accelerometer to confirm that a particular user intended to activate the hot word detection, thereby reducing false detection of other people's voices.

Abstract: A device capable of motion includes an acoustic echo canceller for cancelling a reference signal from received audio data. The device updates an adaptive filter as the device moves to reflect the changing audio channel between a loudspeaker and a microphone of the device. A step size for changing coefficients of the filter is determined based on its velocity. A number of iterations for updating the filter using a frame of audio data is also determined based on the velocity.

Abstract: An active noise cancellation (ANC) system may include an adaptive filter divergence detector for detecting divergence of the one or more controllable filters as they adapt, based on dynamically adapted thresholds. Upon detection of a controllable filter divergence, the ANC system may be deactivated, or certain speakers may be muted. Alternatively, the ANC system may modify the diverged controllable filters to restore proper operation of the noise cancelling system.

Abstract: A method for automatic monitoring of industrial control systems, comprising: monitoring an environment by an electromagnetic sensor to measure a plurality of electromagnetic pulses emitted by at least one industrial control device during operation of the at least one industrial control device; calculating automatically at least one normal data pattern based on an analysis of the plurality of electromagnetic pulses; matching between at least one new electromagnetic pulse measured by the electromagnetic sensor and the at least one normal data pattern to automatically detect at least one abnormal data pattern; and sending automatically a security alert in response to the abnormal data pattern.

Abstract: Systems and methods are provided for generating destructive interference signals for use in active noise cancellation in an interior setting of an urban environment. The system employs a combination of conventional ANC and predictive state estimation to generate destruction interference sound waves using a window pane to counteract urban noises.

Abstract: The techniques described herein improve methods to equip a computing device to conduct automatic speech recognition (“ASR”) in talker-independent multi-talker scenarios. In some examples, permutation invariant training of deep learning models can be used for talker-independent multi-talker scenarios. In some examples, the techniques can determine a permutation-considered assignment between a model's estimate of a source signal and the source signal. In some examples, the techniques can include training the model generating the estimate to minimize a deviation of the permutation-considered assignment. These techniques can be implemented into a neural network's structure itself, solving the label permutation problem that prevented making progress on deep learning based techniques for speech separation. The techniques discussed herein can also include source tracing to trace streams originating from a same source through the frames of a mixed signal.

Abstract: A noise reduction system is provided. The noise reduction system may include a sub-band noise sensor, a plurality of sub-band noise reduction modules, and an output module. The sub-band noise sensor may be configured to detect a noise and generate a plurality of sub-band noise signals in response to the detected noise. Each of the plurality of sub-band noise signals may have a distinctive sub-band of the frequency band of the noise. Each of the sub-band noise reduction modules may be configured to receive one of the sub-band noise signals from the sub-band noise sensor and generate a sub-band noise correction signal for reducing the received sub-band noise signal. The output module may be configured to receive the sub-band noise correction signals and output a noise correction signal for reducing the noise based on the sub-band noise correction signals.

Abstract: An active noise control unit generates a noise canceling sound of a single frequency that cancels engine sound heard by an occupant and outputs the noise canceling sound to a speaker through an amplifier. A second harmonic detector and a third harmonic detector detect a magnitude of a harmonic component of the noise canceling sound included in sound collected by a microphone. A controller controls a frequency of the noise canceling sound generated by the active noise control unit to a fundamental frequency of the engine sound based on a rotation frequency of the engine, and in addition, causes an output level control unit to control a level of the noise canceling sound to be output to the amplifier so that the level of the noise canceling sound becomes higher when the magnitude of the harmonic component exceeds a predetermined threshold value.

Abstract: A method of echo cancellation in hands-free communication is disclosed. The method includes: receiving, via a receive signal processor, a far-end audio signal; providing the far-end audio signal to: an acoustic echo canceller module as a reference signal, and at least one loudspeaker for playback; determining an external gain value associated with the far-end audio signal, the external gain applied to the far-end audio signal downstream of the receive signal processor and prior to playback from the at least one loudspeaker; adjusting at least one parameter of the acoustic echo canceller module based on the external gain value; receiving playback output of the far-end audio signal from the at least one loudspeaker as an input signal to a microphone; and processing the microphone input signal by the adjusted acoustic echo canceller module to produce an echo-cancelled signal.

Abstract: Improved audio processing management in a portable converged communication device is provided. A default mode of audio processing is provided where audio input to a microphone is routed and processed simultaneously to a plurality of different processors in parallel. The default mode is suspended in response to a keyword input to a microphone. The keyword is recognized by a speech recognition engine (ASR). The keyword enables audio processing through one of the plurality of processors while selectively suspending other audio processing through the other processors.

Abstract: A telecommunication device includes an audio signal transmission unit configured to receive an audio signal and to transmit the same to a further telecommunication device. The telecommunication device further includes a signaling unit configured to output a signaling when there is a concern that the audio signal is acoustically understandable for third parties or represents a disturbance for third parties. A further telecommunication device includes an audio signal receive unit configured to receive an audio signal from a further telecommunication device and to output the same acoustically, and a signaling unit configured to output a signaling when there is a concern that the output audio signal is acoustically understandable for third parties or represents a disturbance. The telecommunication devices may be interconnected in a system. Also described are corresponding methods for operation and a computer program.

Abstract: A hearing device configured to be worn at a head of a user. The hearing device includes a vibration sensor to detect a vibration conducted through the user's head to the hearing device and to output a vibration signal including information about the vibration. At least part of the vibration can be caused by an own voice activity of the user. A method of operating the hearing device allows a reliable own voice detection at rather low processing effort. The processor determines a presence of an own voice characteristic in the vibration signal at a vibration frequency associated with the own voice characteristic and identifies the own voice activity based on an identification criterion including the presence of the own voice characteristic in the vibration signal at the associated vibration frequency. The own voice characteristic indicative of part of the vibration can be caused by the own voice activity.

Abstract: A hearing device, e.g. a hearing aid, configured to play sound into an ear canal of a user, the hearing device comprising at least one input transducer (ITq) for picking up sound sq at said at least one input transducer from a sound field S around the hearing device and providing corresponding at least one electric input signal(s) INq representing said sound sq, i=1, . . .

Abstract: An electronic device is disclosed. The electronic device comprises: multiple microphones for receiving audio signals generated by multiple sound sources; a communication unit for communicating with a voice recognition server; and a processor for determining the direction in which each of the multiple sound sources is located with reference to the electronic device, on the basis of the multiple audio signals received through the multiple microphones, determining at least one target sound source among the multiple sound sources on the basis of the duration of the determined direction of each of the sound sources, and controlling the communication unit such that the communication unit transmits, to the voice recognition server, an audio signal of a target sound source from which a predetermined voice is generated among the at least one target sound source.

Abstract: A vibration speaker includes: a housing; a diaphragm covering an open side of the housing and having a rim fixed to the housing; a vibration member installed in the housing to vibrate in a vibration direction of the diaphragm and having a magnet forming a magnetic field; a voice coil installed on an inner surface of the diaphragm and configured to vibrate the diaphragm through interaction with the vibration member; a solenoid driver fixed in the housing and configured to form the magnetic field that controls vibration of the vibration member; an amplifier configured to apply a first output signal to the voice coil and apply a second output signal having a phase different from the first output signal to the solenoid driver; and a controller configured to determine whether to apply the second output signal by comparing a frequency of the first output signal with the resonance frequency.

Abstract: A sound state estimating unit detects surrounding sound at a timing at which a notification to a destination user occurs. A user state estimating unit detects a position of the destination user and positions of users other than the destination user at the timing at which the notification occurs. An output control unit controls output of the notification to the destination user at a timing at which it is determined that the surrounding sound detected by the sound state estimating unit is masking possible sound which can be used for masking in a case where the position of the destination user detected by the user state estimating unit is within a predetermined area.

Abstract: A partial enclosure for inhibiting sound passing into and out of the partial enclosure includes an absorber-barrier or an absorber-barrier-absorber, each made from sound absorbing material and sound barrier material and arranged to form the partial enclosure. The enclosure also includes an adaptive frequency matched sound-masking system. The absorber-barrier or absorbed-barrier-absorber is positioned to block or inhibit unwanted sound from various positions of a source of the unwanted sound, or motion of the source of the unwanted sound. The adaptive frequency matched sound-masking system includes a sound generating device arranged on or in the partial enclosure to emit anti-noise signal to cancel or inhibit the unwanted sound.

Abstract: Disclosed is a deep neural network-based method and apparatus for combining noise and echo removal. The deep neural network-based method for combining noise and echo removal according to one embodiment of the present invention may comprise the steps of extracting a feature vector from an audio signal that includes noise and echo; and acquiring a final audio signal from which both noise and echo have been removed, by using a combined nose and echo removal gain estimated by means of the feature vector and deep neural network DNN.

Abstract: A device and method generate a sound masker to mask sound of the ambient environment. More specifically, spectral characteristics of sound in the ambient environment are determined, where the spectral characteristics are determined in terms of auditory excitation patterns. A database of pre-recorded sounds is searched to identify at least one pre-recorded sound that has spectral characteristics corresponding to the spectral characteristics of the sound in the ambient environment. At least a portion of the identified at least one pre-recorded sound is reproduced to mask the sound in the ambient environment.

Abstract: A system for reducing far-field noise produced by well operations includes a passive sound barrier shielding an area, in which the well operations are performed, in an open-air environment. The system further includes a sound sensor to receive near-field noise from the well operations. The system further includes an analysis module, coupled to the sound sensor, to generate an anti-noise signal. The system further includes active anti-noise generators, coupled to the analysis module, to generate anti-noise, based on the anti-noise signal, that destructively interferes with noise from the well operations outside of the passive sound barrier at a predetermined distance from the passive sound barrier. The analysis module generates the anti-noise signal based on the near-field noise, the predetermined distance, and adjustable positions and orientations of the active anti-noise generators.

Abstract: A motion-sensitive acoustic speaker may include a housing; a transmitter associated with the housing, a receiver associated with the housing, a interface component associated with the housing, and a processing device.

Abstract: A hearing system is adapted to be worn by a user and configured to capture sound in an environment of the user and comprises a) a sensor array comprising M transducers for providing M electric input signals representing said sound and having a known geometrical configuration relative to each other; b) a detector unit for detecting movements over time of the hearing system, and providing location data of said sensor array at different points in time t, t=1, . . . , N; c) a first processor for receiving said electric input signals and—in case said sound comprises sound from a localized sound source S—for extracting sensor array configuration specific data ?ij of said sensor array indicative of differences between a time of arrival of sound from said localized sound source S at said respective input transducers, at said different points in time t, t=1, . . .

Abstract: An information handling system audio system includes a library of noise reduction filters associated with cooling fan speeds to isolate out cooling fan noise. Changed cooling fan settings communicated to the audio system trigger application of a library noise reduction filter for the selected cooling fan setting to isolate out cooling fan noise while an adaptive filter defines a noise reduction filter from recorded sounds. The library noise reduction filter reduces cooling fan noises during the time used to determine the adaptive noise reduction filter. In one embodiment, changes in cooling fan settings prioritize definition of noise reduction filters by the adaptive filter.

Abstract: An audio signal processing apparatus for rendering an input audio signal is disclosed. The audio signal processing apparatus includes a receiving unit, which obtains a plurality of input audio signals corresponding to sounds collected by each of a plurality of sound collecting devices, a processor, which obtains an incidence direction for each frequency component for at least some frequency components of each of the plurality of input audio signals corresponding to a sound incident to each of the plurality of sound collecting devices based on cross-correlations between the plurality of input audio signals, and generates an output audio signal by rendering at least some of the plurality of input audio signals based on the incidence direction for each frequency component, and an output unit, which outputs the generated output audio signal.

Abstract: A wireless earpiece includes a wireless earpiece housing, a processor disposed within the wireless earpiece housing, at least one microphone operatively connected to the processor, and at least one speaker operatively connected to the processor. The processor is configured to receive audio from the at least one microphone, perform processing of the audio to provide processed audio, and output the processed audio to the at least one speaker. The processing of the audio involves identifying body generated sounds generated by a body of a user of the wireless earpiece and removing the body generated sounds.

Abstract: A method for removing a target unwanted signal from multiple signals. The method includes: providing a set of input signals from external devices; separating the input signals into channels with the unwanted signal and channels without the unwanted signal; synchronizing the sets of input signals; and transferring the separated signal via wire or wirelessly to a sound reproduction device.

Abstract: A loudspeaker assembly comprising: a first loudspeaker configured to receive a first electrical signal, and to produce sound along a first principal radiating axis based on the first electrical signal; a second loudspeaker configured to receive a second electrical signal, and to produce sound along a second principal radiating axis based on the second electrical signal; a third loudspeaker configured to receive a third electrical signal, and to produce sound along a third principal radiating axis based on the third electrical signal; and a control unit configured to produce each of the first, second and third electrical signals based on an input signal representative of audio. There is a first angular offset between the first and second principal radiating axes and a second angular offset between the first and third principal radiating axes.

Abstract: Voice responsive in-wall devices are provided. In one example implementation, a power switch includes a housing mountable on or at least partially within a surface. The housing can have a front panel. The power switch can include an interface element disposed on the front panel and operable to receive a user input. The power switch can include a power interrupter operable to control power delivery to the powered load based at least in part on interaction with the interface element. The power switch can include one or more microphones operable to obtain audio input. The power switch can include one or more speakers configured to provide audio output. The power switch can include a communications interface operable to communicate data associated with the audio input over a communication link.

Abstract: A system and method for constructing training dictionaries with multichannel information. An exemplary method takes into account the effect of the acoustic path while training multichannel acoustic data. A method that uses different time-frequency resolutions in machine learning training is also presented.

Abstract: Embodiments of the present invention provide a signal processing method and apparatus. The method includes: performing M-way filtering on an input signal to obtain M filtered signals, performing extraction on M filtered signals separately to obtain M extracted signals, performing fast Fourier transform (FFT) on the M extracted signals separately to obtain M frequency-domain signals, and finally determining output signals according to the M frequency-domain signals. According to the embodiments of the present invention, signal filtering and extraction are performed and then FFT is performed.

Abstract: An improved system and method for recognizing an audio signal due to physical activity and taking a predetermined action in response is disclosed. A “reverse noise signal” created by the sound pressure wave of the physical activity acting on the earpiece transducer is obtained. In some embodiments, an ambient noise signal is inverted and fed back, and the inverted signal is added to the intended audio signal being sent to the earpiece so that the ambient noise is cancelled. In other embodiments, a processor receives the ambient noise signal and predicts the modification to the intended audio signal needed to counteract the ambient noise. In other embodiments, the reverse noise signal may represent a motor or biological activity of a user; the system may take different actions in response to different physical activities, such as a heart beat of the user, or a tap, footfall, or swallowing by the user.

Abstract: In a vehicle equipped with a vehicle sound system, determining a virtual fixed gear ratio; determining a virtual RPM based on the virtual fixed gear ratio; generating a set of harmonic signals based on the virtual RPM, the harmonic signals being sine-wave signals proportional to harmonics of the virtual RPM; processing the harmonic signals to produce a set of processed harmonic signals; and in the vehicle sound system, transducing the processed harmonic signals to acoustic energy thereby to produce an engine sound within a passenger cabin of the vehicle.

Abstract: A method, apparatus and computer program including: obtaining a spatial audio signal from a plurality of microphones; dividing the obtained spatial audio signal into at least a first component and a second component; applying a first audio signal optimizing system to the first component and applying a second audio signal optimizing system to the second component; and enabling a signal including the optimized components to be provided to a speaker for rendering.

Abstract: In an approach to adaptive sound masking, one or more computer processors analyze a surrounding of one or more users and stores in a database. The one or more computer processors receive a request from the one or more users for adaptive sound masking. The one or more computer processors analyzes a surrounding environment associated with the one or more users and storing a first information associated with the surrounding environment in a database. The one or more computer processors generate a cognitive sound mask base on the first information. The one or more computer processors produce a sound cone based on the cognitive sound mask and directing the sound cone at a distracting sound. The one or more computer processors adapt the sound cone based on changes to the surrounding environment.

Abstract: The present disclosure relates to systems and methods for active sound management for vehicles. In one embodiment, a process for active sound management for a vehicle includes determining vehicle operation state and powertrain operation for the vehicle, and detecting sound generated by the powertrain operation of the vehicle. A synthesized sound may be determined for output based on selection of at least one frequency of the sound generated by the powertrain to drive a frequency of a sine-wave generator, synthesizing an output wave file with pitch shifting of frequency based on powertrain operation, and blending sine-wave generator output with the synthesized wave file to cancel unwanted powertrain noise and generate desired sounds for vehicles. Output of the synthesized sound may be controlled to provide simultaneous cancellation and synthesis in at least one same frequency range.

Abstract: [Object] To provide a sound processing device capable of effectively reducing ambient noise at low cost. [Solution] Provided is a sound processing device including: a first sound collector configured to collect a first noise signal from a noise source of noise leaking into a casing mounted to a user's ear; a first signal processing unit configured to form a first noise reduction signal used to reduce noise at a cancellation point on the basis of the first noise signal; a second signal processing unit configured to form a second noise reduction signal used to reduce noise at a cancellation point with respect to a first pseudo noise signal; an adder configured to add the first noise reduction signal and the second noise reduction signal; and a sound emitter configured to emit an output of the adder into the casing as sound.

Abstract: Apparatus and methods for processing an audio signal. The apparatus may include a digital signal processor (“DSP”) configured to receive an audio signal from a vehicular audio signal line. The audio signal may include a flat component and a compensatory component. The flat component may correspond to the audio signal in a state before combination with the compensatory component. The compensatory component may include an OEM EQ component. The apparatus may include a microprocessor in electronic communication with the DSP. The microprocessor may be configured to retrieve from the memory a restorative signal component. The microprocessor may be configured to instruct the DSP to apply the restorative signal component to the audio signal to reduce the audio signal to the flat component.

Abstract: An example system includes non-transitory machine-readable storage storing calibration data sets. A calibration data set includes parameter values that vary non-linearly. Each of the calibration data sets is temperature-specific. The example system also includes channels over which signals pass to and from units under test (UUTs). A channel includes input circuitry to receive a signal of the signals and to obtain a first parameter based on the signal; and correction circuitry to obtain a second parameter based on the first parameter and based on the calibration data set. The second parameter includes a calibrated version of the first parameter. The calibration data set is selectable based on temperature.

Abstract: Active noise cancellation (ANC) systems and methods include a reference sensor to sense external noise and generate a reference signal, an error sensor to sense noise in a noise cancellation zone and generate an error signal, a feedforward ANC subsystem to receive the reference and error signals and generate a first anti-noise signal to cancel external noise in a noise cancellation zone, a feedback ANC subsystem to receive the error signal and generate a second anti-noise signal to cancel the external noise in the cancellation zone, a wind detector to detect whether wind noise is present in the reference signal and output a wind noise detection status, and wind handler to control adaptation processing of the feedforward ANC subsystem and the feedback ANC subsystem in accordance with the wind noise detection status and mixing of the first and second anti-noise signals to generate an output anti-noise signal.

Abstract: Methods and systems provide dynamic selection of noise-cancelling algorithms, and dynamic activation and deactivation of microphones to provide multi-mode noise cancellation for a voice-detecting headset in situations where ambient noise prevents voice navigation from accurately interpreting voice commands. To do so, when an ambient noise is detected that exceeds a threshold, a particular noise-cancelling algorithm best-suited for the situation is selected, and one or more noise-detecting microphones is activated. The noise-detecting microphone(s) receiving the highest level of ambient noise can remain activated while the remaining noise-detecting microphones can be deactivated. A speech signal received by the speech microphone can then be optimized by cancelling the ambient noise signal received from the activated noise-detecting microphone(s) using the selected noise-cancelling algorithm. After the speech signal is optimized, it can be communicated to the voice-detecting headset for interpretation.

Abstract: Active noise control system and methods are disclosed that include generating anti-noise that is configured to reduce or cancel noise occurring at a listening position in the ambiance of a seat with a backrest enclosing a hollow volume. Radiating the anti-noise from within the hollow volume of the backrest of the seat.

Abstract: In a noise reduction device that generates and outputs a control sound signal for reducing noise, an internal loop control unit controls an internal loop in which a pre-output control sound signal that is acquired from a control sound output unit before output to a speaker is input to a sound receiver. A measurement unit measures an input level of a microphone sound signal and an input level of the pre-output control sound signal that has been input to the sound receiver in the internal loop. A fault detector uses the input level of the microphone sound signal and the input level of the pre-output control sound signal measured by the measurement unit to detect a fault in any one of the microphone, the sound receiver, the speaker, and the control sound output unit. A transmitter sends a result of fault detection performed by the fault detector to a management device.

Abstract: A method includes accessing, by at least one processing device, an audible signal including at least one in-ear microphone audible signal and at least one external microphone audible signal and at least one noise signal; training a generative network to generate an enhanced external microphone signal from an in-ear microphone signal based on the at least one in-ear microphone audible signal and the at least one external microphone audible signal; and outputting the generative network.

Abstract: Systems and methods are provided for an active noise cancellation system which employs a combination of an active noise cancellation techniques and predictive state estimation to generate destructive interference audio signals that are propagated by an interior surface of a room to counteract urban noises in an exterior environment of the room.

Abstract: An apparatus comprising an audio detector configured to analyse a first audio signal to determine at least one audio source, wherein the first audio signal is generated from the sound-field in the environment of the apparatus; an audio generator configured to generate at least one further audio source; and a mixer configured to mix the at least one audio source and the at least one further audio source such that the at least one further audio source is associated with the at least one audio source.