Starkey Laboratories, Inc.

Fraunhofer-Gesellschaft Zur Foerderung Der Angewandten Forschung E.V.

Henning Schepker

Wenyu Jin

Joyce Rosenthal

Christophe Daniel Micheyl

Jumana Harianawala

Martin McKinney

Jens Ekkehart Appell

Majid Mirbagheri

Jan Rennies

Simon Doclo

Sina Miran

Lior Weizman

Liron Pollak

Masahiro Sunohara

Piero Iared Rivera Benois

Disclosed herein, among other things, are systems and methods for active noise cancellation (ANC) for hearing device applications. A method includes estimating a first sound pressure on an ear drum of a wearer of the hearing device caused by a hearing processing signal, estimating a second sound pressure on the ear drum of the wearer of the hearing device caused by a leakage path of the hearing device, and computing a ratio of the first sound pressure and the second sound pressure to predict a comb-filtering effect for the hearing device. The method also includes computing an ANC controller using the computed ratio in one or more frequency ranges, and canceling leaked sound into an ear canal of the wearer for the hearing device in the one or more frequency ranges using the ANC controller.

Active noise control for sound quality in hearing devices

A hearing device includes a deep/recurrent neural network trained to jointly perform sound enhancement and feedback cancellation. During training a neural network is connected between a simulated input and a simulated output of the hearing device. The neural network is operable to change a response affecting the simulated output. The neural network is trained by applying the simulated input to the deep neural network while applying the feedback path response between the simulated input and the simulated output. The deep-neural network is trained to reduce an error between the simulated output and the reference audio signal and used for sound enhancement in the device.

Apparatus and method for speech enhancement and feedback cancellation using a neural network

A hearing device includes a microphone that produces an audio input signal and a loudspeaker that outputs an amplified audio signal into an ear canal. A signal processing path is coupled to the microphone and the loudspeaker. The signal processing path includes a deep neural network configured to predict an instantaneous gain margin of the hearing device based on a set of inputs. The set of inputs includes a first parameter of the audio input signal, a second parameter of the amplified audio signal, and a gain of the signal processing path. A feedback reduction module of the device receives the predicted instantaneous gain margin and adjusts feedback reduction parameters to reduce an onset of feedback in the hearing device.

Predicting gain margin in a hearing device using a neural network

Disclosed herein, among other things, are systems and methods for artifact detection and tuning of a feedback canceller. A method includes detecting an artifact of a feedback canceller during operation of a hearing device. Artifact data associated with the detected artifact is stored in a memory of the hearing device, the artifact data including information related to the detected artifact and information related to the hearing device during occurrence of the detected artifact. Artifact statistics are computed based on the stored artifact data, and the artifact data and the artifact statistics are transmitted to an external device. Comparison data is received from the external device, including results of a comparison of artifact data or artifact statistics of the hearing device to artifact data or artifact statistics of other devices. Parameters of the hearing device are automatically adjusted based on the comparison data to improve performance of the feedback canceller.

Artifact detection and logging for tuning of feedback canceller

An individual cutoff frequency for an individual secondary path measurement is determined. First and second acoustic transducer-to-eardrum responses below and above the cutoff frequency are determined using an individual secondary path measurement, a reduced dimensionality estimate, and a least squares estimate. A sound pressure level at an eardrum of the user is predicted using the first and second receiver-to-eardrum responses.

Apparatus and method for estimation of eardrum sound pressure based on secondary path measurement

Disclosed herein are systems and methods for eardrum acoustic pressure estimation for hearing device applications. A receiver signal is produced through a receiver of a hearing device, or a physical extension of the receiver, placed inside an ear canal of a user. A microphone signal is sensed, in response to the receiver signal, using a microphone placed outside the ear canal. Based on the receiver signal and the microphone signal, a reliability signal of the microphone signal and a feedback-path signal are computed. A spectral representation signal of the feedback-path signal is computed, and a real-ear response signal is computed using a mathematical combination of the spectral-representation signal and the reliability signal. The computed real-ear response signal is used to modify an operational characteristic of the hearing device, or to make a recommendation for modification of an operational characteristic or a physical characteristic of the hearing device.

Eardrum acoustic pressure estimation using feedback canceller

Secondary path measurements and associated acoustic transducer-to-eardrum responses are obtained from test subjects. Both a least squares estimate and a reduced dimensionality estimate are determined that both estimate a relative transfer function between the secondary path measurements and the associated acoustic transducer-to-eardrum responses. An individual secondary path measurement for a user is performed based on a test signal transmitted via a hearing device into an ear canal of the user. An individual cutoff frequency for the individual secondary path measurement is determined. First and second acoustic transducer-to-eardrum responses below and above the cutoff frequency are determined using the individual secondary path measurement and the least squares estimate. A sound pressure level at an eardrum of the user can be predicted using the first and second receiver-to-eardrum responses.

An apparatus for generating a modified speech signal from a speech input signal which has a plurality of speech subband signals, the modified speech signal having a plurality of modified subband signals is provided, having: a weighting information generator for generating weighting information for each speech subband signal depending on a signal power of said speech subband signal, and a signal modifier for modifying each speech subband signal by applying the weighting information on said speech subband signal to obtain a modified subband signal. The weighting information generator is configured to generate the weighting information for each of the plurality of speech subband signals, wherein the signal modifier is configured to modify each of the speech subband signals so that a first speech subband signal having a first signal power is amplified with a first degree, and so that a second speech subband signal having a second signal power is amplified with a second degree, the first signal power being greater than the second signal power, and the first degree being lower than the second degree.

Growing community of inventors

Oldenburg, Germany

Henning Schepker