An active noise reduction device includes: an adaptive filter that generates a cancelling signal by applying an adaptive filter to a reference signal that has a correlation with noise in a space in an automobile, the cancelling signal being used to output a cancelling sound for reducing the noise; and a filter coefficient updater that calculates a coefficient of the adaptive filter based on a predetermined update equation. At a first timing at which the output of the cancelling sound is started, the filter coefficient updater uses a first coefficient as an initial value of the update equation, the first coefficient being the coefficient of the adaptive filter calculated by the filter coefficient updater at a second timing that is prior to the first timing.

An active noise control device includes a basic signal generating unit configured to generate a basic signal corresponding to a resonance frequency of a vibration sensor, a first adaptive filter configured to generate a sensor resonance simulation signal simulating a signal acquired while the vibration sensor is resonating by performing a filtering process on the basic signal, a computation unit configured to calculate a second reference signal that is a difference between a first reference signal acquired by the vibration sensor and the sensor resonance simulation signal, and a second adaptive filter configured to generate a control signal by performing a filtering process on the second reference signal.

An active noise control device includes a determination unit configured to determine whether an abnormality has occurred in an acceleration sensor based on a direct-current component of a reference signal acquired by the acceleration sensor attached to a vehicle, and a control unit configured to, when the determination unit determines that an abnormality has occurred in any of a plurality of the acceleration sensors, stop generation of the control signal based on the reference signal acquired by the acceleration sensor that has been determined to have the abnormality, and stop updating a filter coefficient of an adaptive filter configured to perform the filtering process on the reference signal acquired by the acceleration sensor that has been determined to have no abnormality.

A method for eliminating unstable noise is provided and applicable to a sound recording device and implemented by a codec. The method includes: activating the sound recording device to start recording; setting a suppression duration and a cutoff frequency switching duration according to unstable noise and a DC offset value of the sound recording device; processing a front-end audio of a recorded sound by a filter having a first cutoff frequency to make the unstable noise in the front-end audio quickly converge, and outputting a filtered audio signal; suppressing the filtered audio signal according to the suppression duration to eliminate the unstable noise; and adjusting the first cutoff frequency of the filter to a second cutoff frequency according to the cutoff frequency switching duration, where the first cutoff frequency is greater than the second cutoff frequency. A device for eliminating unstable noise is also provided.

An active noise reduction device includes: a reference signal inputter; an adaptive filter; a μ adjuster that calculates a step size parameter by multiplying a reference value for the step size parameter by a correction coefficient that is proportional to a reciprocal of a first representative input value that indicates a signal level of the reference signal in a first predetermined period; a filter coefficient updater that updates adaptive filter coefficient W by using the step size parameter calculated; and a determiner. When it is determined that a second representative input value is greater than a threshold value, at least one of the adaptive filter or the filter coefficient updater is transitioned from a normal state to a restriction state in which an effect of reducing noise is smaller than in the normal state.

An active noise reduction device includes a first adaptive filter, a first filter coefficient updater, and a controller that determines, based on a first parameter of the first adaptive filter, whether first noise control based on a first cancelling sound is in a stable state or an unstable state. The controller transitions the first filter coefficient updater to a restriction state in which an effect of reducing first noise is smaller than in a normal state when it is determined that the first noise control is in the unstable state while the first filter coefficient updater is updating a coefficient of the first adaptive filter in the normal state, and transitions the first filter coefficient updater back to the normal state when it is determined that the first noise control is in the stable state while the first filter coefficient updater is in the restriction state.

An illustrative controller includes: a transmitter to drive the acoustic transducer to generate acoustic bursts; a receiver to sense a response of the acoustic transducer to echoes; and a processing circuit coupled to the transmitter and to the receiver, the processing circuit configured to convert said received response into output data by: correlating said response to a driving signal to obtain a correlation response; distinguishing peak areas from non-peak areas in said correlation response; deriving a noise level in a portion of said correlation response based on the non-peak areas within said portion; calculating a signal to noise ratio (SNR) for a peak signal within the portion as a ratio of a peak value for the peak signal to the noise level in said portion of said correlation response; and accepting the peak signal as an echo only if the SNR for said peak signal exceeds a predetermined threshold.

A cooperative control unit is configured to cause a muffled sound control signal processing unit to start feedforward signal processing, and set, when a secondary path filter has converged, a narrow band noise control signal processing unit to the converged secondary path filter and cause the narrow band noise control signal processing unit to start feedback signal processing.

A method performed by an in-ear headphone that includes a speaker and an internal microphone. The method receives a microphone signal from the internal microphone that indicates a current sound pressure level (SPL) in an ear canal of a user. The current SPL is a result of a control leak from the in-ear headphone into an ambient environment that reduces a SPL in the ear canal between 2 dB and 25 dB at a frequency within a frequency range than if otherwise not present. The method determines an active noise cancellation (ANC) filter based on the microphone signal and generates an anti-noise signal using the ANC filter. The method drives the speaker using the anti-noise signal to reduce the current SPL in the ear canal of the user as much as 25 dB at a frequency within the frequency range.

The disclosure provides a method and a system for head-related transfer function (HRTF) adaptation. The method includes performing a system identification. The system identification includes a pinna identification and a shadowing identification.