The handling of disturbances to audio signals may be improved with an adaptive noise cancellation (ANC) system that performs frequency-domain adaption. The ANC systems may be configured to determine if a disturbance is present at a first frequency in the second input signal received from the reference microphone. The ANC systems may update an algorithm of an adaptive filter based, at least in part, on the first input signal, the second input signal, and a feedback signal that is based on an output of the adaptive filter by changing parameters of the algorithm such that the adaptive filter adapts around the first frequency differently than other frequencies when the disturbance is present.

An audio processing scheme is described. In an example, an apparatus comprises: at least two acoustic sensors through which audio content is received; at least one other sensor; an audio processor connected to the sensors and configured to receive audio information from the acoustic sensors and other information from the other sensor. The audio processor is configured to determine a use case of the apparatus based on the audio information and the other information. The audio processor is configured to adjust at least one audio processing scheme for the apparatus based on the determined use case. In other examples, a method and a computer program product are described.

An integrated circuit for implementing at least a portion of a personal audio device may include a processing circuit to implement an adaptive filter having a response that generates an anti-noise signal to reduce the presence of the ambient audio sounds at an error microphone, implement a coefficient control block that shapes the response of the adaptive filter in conformity with the error microphone signal by computing coefficients that determine the response of the adaptive filter to minimize the ambient audio sounds at the error microphone, and responsive to detecting a condition that triggers a reset of the adaptive filter, increment the coefficients in a plurality of steps from initial values of the coefficients at a time of triggering the reset to final values of the coefficients at a conclusion of the reset.

A personal audio device, such as a wireless telephone, includes noise canceling that adaptively generates an anti-noise signal from a reference microphone signal and injects the anti-noise signal into the speaker or other transducer output to cause cancellation of ambient audio sounds. An error microphone is provided proximate the speaker to measure the output of the transducer in order to control the adaptation of the anti-noise signal and to estimate an electro-acoustical path from the noise canceling circuit through the transducer. The anti-noise signal is adaptively generated to minimize the ambient audio sounds at the error microphone. A processing circuit that performs the adaptive noise canceling (ANC) function also filters one or both of the reference and/or error microphone signals, to bias the adaptation of the adaptive filter in one or more frequency regions to alter a degree of the minimization of the ambient audio sounds at the error microphone.

The handling of disturbances to audio signals may be improved with an adaptive noise cancellation (ANC) system that performs tone suppression and howl suppression in a collaborative manner. Such ANC systems may be configured to detect a first tone in an input signal at a first tone frequency and extract the detected first tone from the input signal. The ANC systems may also be configured to adaptively filter the extracted first tone to generate a second tone that has a magnitude that is approximately equal to a magnitude of the extracted first tone and a phase that is approximately opposite the phase of the extracted first tone. The ANC systems may be further configured to add the second tone to an intermediate signal that is based, at least in part, on the input signal to generate the output signal.

A system for self-organized acoustic signal cancellation over a network is disclosed. The system may transmit an acoustic sounding signal to an interfering device so that a channel measurement may be performed for a channel between the interfering device and an interferee device. The system may receive the channel measurement for the channel from the interfering device and also receive a digitized version of an audio interference signal associated with the interfering device. Based on the channel measurement and the digital version of the interference signal, the system may calculate a cancellation signal prior to the arrival of the original over-the-air audio interference signal that corresponds to the digital version of audio interference signal. The system may then apply the cancellation signal to an audio signal associated with the interferee device to remove the interference signal from the audio signal.

An audio processing system includes: a first microphone configured to output a first signal based on a first audio signal; one or more microphones each of which outputs a microphone signal based on an audio signal; one or more adaptive filters configured to respectively receive the microphone signals from the one or more microphones and output passing signals based on the microphone signals; and a processor configured to: determine whether the microphone signal includes uncorrelated noise; control one or more filter coefficients of the one or more adaptive filters; and subtract a subtraction signal based on the passing signals from the first signal. The one or more microphones include a second microphone that outputs a second signal. When determining that the second signal includes the uncorrelated noise, the processor is configured to set a level of the second signal input to the corresponding adaptive filter to zero.

An acoustic echo canceller (AEC) uses a simulated echo from an adaptive filter to cancel an actual echo from a real environment. The echo cancellation is optimal when the adaptive filter response accurately matches the environment response. This can be achieved by changing coefficients of the filter in a step-wise fashion until the filter is converged on a configuration that provides optimal results. The convergence of the filter can be negatively affected by interference due to a double talk event or by an event in which the environment is changed. The disclosed AEC is configured to respond to these disruptive events by adjusting a step-size of the adaptive algorithm used to change the coefficients.

Embodiments disclosed herein include security systems and methods for securing an intelligent automated assistant device comprising. In some embodiments, the security system may include an intelligent automated assistant device including a microphone and a camera. Additionally, the security device may be placed near the intelligent automated assistant device. The security device may also include security components to distort sounds from a sound source to be detected by the microphone. As a result, this may prevent third parties from at least remotely streaming or recording live audio from a microphone on the intelligent automated assistant device.

A filtering device includes a low-pass filter (LPF), a noise estimation circuit and a first combining circuit. The LPF receives and filters a pre-filtering signal to generate an output signal of the filtering device. The noise estimation circuit estimates an estimated noise signal according to the output signal and the pre-filtering signal. The first combining circuit subtracts the estimated noise signal from an input signal of the filtering device to generate the pre-filtering signal.