An assistive listening device includes a set of microphones including an array arranged into pairs about a nominal listening axis with respective distinct intra-pair microphone spacings, and a pair of ear-worn loudspeakers. Audio circuitry performs arrayed-microphone short-time target cancellation processing including (1) applying short-time frequency transforms to convert time-domain audio input signals into frequency-domain signals for every short-time analysis frame, (2) calculating ratio masks from the frequency-domain signals of respective microphone pairs, wherein the calculation of a ratio mask includes both a frequency domain subtraction of signal values of a microphone pair and a scaling of a resulting frequency domain noise estimate by a pre-computed phase difference normalization vector, (3) calculating a global ratio mask from the plurality of ratio masks, and (4) applying the global ratio mask, and inverse short-time frequency transforms, to selected ones of the frequency-domain signals, thereby generating audio output signals for driving the loudspeakers. The circuitry and processing may also be realized in a machine hearing device executing a human-computer interface application.

A method for fast recognition of a hearing aid wearer's own voice, and a corresponding apparatus for carrying out the method. The hearing aid receives audio signals from at least two acoustoelectric transducers. The hearing aid has an apparatus with a first filter and a second filter for spatial separation. First filter parameters of the first filter are ascertained and used to attenuate the own voice of the hearing aid wearer of the hearing aid. In addition, second filter parameters of the second filter are ascertained. The second filter parameters are used to attenuate an external audio source. The apparatus is then operated with the first and second parameters, and the wearer's own voice is recognized on the basis of an output signal from the first filter and the second filter.

A hearing device adapted for being worn at or in an ear of a user, comprises a) an input unit comprising at last two input transducers each for converting sound around said hearing device to an electric input signal representing said sound, thereby providing at least two electric input signals; b) a beamformer filter comprising a minimum processing beamformer defined by optimized beamformer weights, the beamformer filter being configured to provide a filtered signal in dependence of said at least two electric input signals and said optimized beamformer weights; c) a reference signal representing sound around said hearing device; d) a performance criterion for said minimum processing beamformer. The minimum processing beamformer is a beamformer that provides the filtered signal with as little modification as possible in terms of a selected distance measure compared to said reference signal, while still fulfilling said performance criterion. The optimized beamformer weights are adaptively determined in dependence of said at least two electric input signals, said reference signal, said distance measure, and said performance criterion. A method of operating a hearing device is further disclosed. The invention may e.g. be used in hearing aids or headsets.

An illustrative stereo rendering system obtains a contradirectional audio input signal generated by a microphone assembly having a plurality of microphone elements. The contradirectional audio input signal implements a contradirectional polar pattern oriented with respect to a listener. The system also obtains an array of multidirectional audio input signals generated by the microphone assembly. The array of multidirectional audio input signals implements different unidirectional polar patterns that are collectively omnidirectional in a horizontal plane. The system generates a weighted audio input signal by mixing the array of multidirectional audio input signals in accordance with respective weight values assigned to each multidirectional audio input signal. The system then generates, based on the contradirectional audio input signal and the weighted audio input signal, a stereo audio output signal for presentation to the listener. Corresponding systems and methods are also disclosed.

A hearing aid configured to be worn by a user, comprises a) an input unit configured to provide an electric input signal representing sound, b1) a wake word detector configured to identify a particular wake word based on said electric input signal, or a signal derived therefrom, and to provide a wake word control signal indicative of whether or not, or with what probability, said wake word is detected, or b2) an own voice detector configured to estimate whether or not, or with what probability, the electric input signal or a signal derived therefrom originates from the voice of the user and to provide an own voice control signal indicative thereof, c) transceiver circuitry configured to establish a communication link to another hearing aid allowing the transmission of said electric input signal, or a signal derived therefrom, to said another hearing aid and/or the reception of an electric input signal, or a signal derived therefrom, from said another hearing aid., and d) a pre-processor configured to control said transceiver circuitry in dependence of said wake word control signal or in dependence of said own voice control signal. A binaural hearing aid system and a method of operating a hearing aid are further disclosed. The invention may e.g. be used in binaural hearing aid systems and associated applications, e.g. communication systems.

A hearing device, e.g. a hearing aid or a headset, configured to be worn by a comprises an input unit for providing at least one electric input signal in a time-frequency representation; and a signal processor comprising a target signal estimator for providing an estimate of the target signal; a noise estimator for providing an estimate of the noise; and a gain estimator for providing respective gain values in dependence of said target signal estimate and said noise estimate. The gain estimator comprises a trained neural network, wherein the outputs of the neural network comprise real or complex valued gains, or separate real valued gains and real valued phases. The signal processor is configured—at a given time instance t—to calculate changes Δx(i,t)=x(i,t)−{circumflex over (x)}(i,t−1), and Δh(j,t−1)=h(j,t−1)−ĥ(j,t−2) to an input vector x(t) and to the hidden state vector h(t−1), respectively, from one time instance, t-1, to the next, t, and where {circumflex over (x)}(i,t−1) and ĥ(j,t−2) are estimated values of x(i,t−1) and h(j,t−2), respectively, where indices i, j refers to the i.sup.th input neuron and the j.sup.th neuron of the hidden state, respectively, where 1≤i≤N.sub.ch,x and 1≤j≤N.sub.ch,oh, wherein N.sub.ch,x and N.sub.ch,oh is the number of processing channels of the input vector x and the hidden state vector h, respectively, and wherein the signal processor is further configured to provide that the number of updated channels among said N.sub.ch,x and said N.sub.ch,oh processing channels of the modified gated recurrent unit for said input vector x(t) and said hidden state vector h(t−1), respectively, at said given time instance t is limited to a number of peak values N.sub.p,x, and N.sub.p,oh, respectively, where N.sub.p,x is smaller than N.sub.ch,x, and N.sub.p,oh, is smaller than N.sub.ch,oh.

A hearing aid system for assisting a user's ability to hear includes at least one hearing aid instrument worn on the user's head. A sound signal from the surroundings is recorded and converted into input audio signals by input transducers of the hearing aid system. The hearing aid system includes two adaptive beamformers with variable notch direction, applied indirectly or directly to the input audio signals to generate direction-dependently damped audio signals. The notch directions are set to mutually different values to minimize the energy content of the direction-dependently damped audio signal of each beamformer. The notch directions of the two beamformers are evaluated in comparative fashion. A user's head rotation is captured qualitatively and/or quantitatively if a correlated change in the notch directions is determined within the scope of the comparative evaluation. A method for operating the hearing aid system is also provided.

In a method for direction-dependent noise rejection for a hearing system, first and second input transducers are used to generate an interference signal and a target signal from a sound from the surroundings. The interference signal and/or the target signal are referenced to a useful signal source arranged in a target direction. The target signal is generated with a target directivity pattern. For each of a first plurality of frequency bands, an acoustic characteristic of the target signal is compared with a corresponding acoustic characteristic of the interference signal, and the comparison is used to ascertain a provisional weighting factor. The provisional weighting factor is used to form for the frequency band a weighting factor for the respective frequency bands. An input signal to be processed is weighted on a frequency-band-by-frequency-band basis using the respective weighting factor, and the weighted input signal is used to generate an output signal.

A hearing aid system assists a user's ability to hear. The system has a hearing aid instrument worn on the user's head. A sound signal from the user's surroundings is recorded and converted into input audio signals by two input transducers. The input audio signals are processed in a signal processing step for generating an output audio signal, which is output by an output transducer. The input audio signals or audio signals derived therefrom by pre-processing are direction-dependently damped by an adaptive beamformer according to the stipulation of a variable directivity with a directional strength to generate a directed audio signal. The directivity is varied with a specified adaptation speed such that the energy content of the directed audio signal is minimized. The adaptation speed and/or the directional strength are variably set on a basis of an analysis of the input audio signals or of the pre-processed audio signals.

A method operates a hearing instrument that is worn in or at the ear of a user. The method includes capturing a sound signal from an environment of the hearing instrument; analyzing the captured sound signal to recognize own-voice intervals, in which the user speaks, and foreign-voice intervals, in which at least one different speaker speaks; and determining, from the recognized own-voice intervals and foreign-voice intervals, at least one turn-taking feature. From the at least one turn-taking feature a measure of the sound perception by the user is derived. Predefined action for improving the sound perception is taken if the measure or the at least one turn-taking feature fulfill a predefined criterion.