Cirrus Logic Inc.

Rahim Saeidi

Pablo Peso Parada

John L Melanson

Sara Mainiero

Toby Stokes

Embodiments of the present disclosure provide apparatus, methods and computer programs for ultrasonic proximity-detection. In one embodiment, processing circuitry comprises: a first input for receiving an indication of an interrogating ultrasonic signal; a second input for receiving a detected ultrasonic return signal; an adaptive filter, coupled to the first and second inputs, operative to estimate a transfer function between the interrogating ultrasonic signal and the detected ultrasonic return signal; a feature extract module, coupled to the adaptive filter, operative to calculate statistical moments of one or more of: the estimated transfer function; an estimated ultrasonic return signal, calculated based on the interrogating ultrasonic signal and the estimated transfer function; and an error signal, representing the error between the estimated ultrasonic return signal and the detected ultrasonic return signal; and a classifier module, coupled to the feature extract module, operative to determine the presence of a nearby object based on the statistical moments.

Proximity sensing

A howling detector is described that is configured to receive an input signal and to determine measure of the linearity of a logarithmic representation of the energy of the input signal. In some examples, this triggers gain adjustment (e.g. of a noise control unit) and, in some further examples, the amount of the gain adjustment may be based on an estimation of the maximum stable gain of a noise control unit.

Techniques for howling detection

An audio processing system is described for an audio event detection (AED) system. The system includes a feature extraction block configured to derive at least one feature which represents a spectral feature of the input signal.

Sound event detection

A method of speaker modelling for a speaker recognition system, comprises: receiving a signal comprising a speaker's speech; and, for a plurality of frames of the signal: obtaining a spectrum of the speaker's speech; generating at least one modified spectrum, by applying effects related to a respective vocal effort; and extracting features from the spectrum of the speaker's speech and the at least one modified spectrum. The method further comprises forming at least one speech model based on the extracted features.

Speaker enrollment

A processing module for a noise control circuit may include a howling detector configured to receive an input signal and to determine a linearity metric based on the input signal, the linearity metric comprising a measure of the linearity of a logarithmic representation of the energy of the input signal. The processing module may also include a gain adjuster configured to adjust the gain of a noise control unit.

The present application describes techniques for noise control which utilize a feedback control unit comprising a filter, derived from one or more predetermined filter candidates, for reducing or cancelling a feedback component of a noise control signal.

Growing community of inventors

Surrey, United Kingdom

Rahim Saeidi