Hearing device comprising an own voice estimator
Disclosed herein are embodiments of a hearing device including at least one first, outward-facing, input transducer configured to pick up first sounds from the environment of a user and a second, inward-facing, input transducer configured to pick up a second sounds at the eardrum of the user. The hearing device can further include a directional system including a) an own voice beamformer configured to provide an estimate of the user's own voice in dependence of the at least one first and the second electric input signals and configurable own voice beamformer weights; and b) an own voice analyzer configured to analyze at least one of the at least one first and said second electric input signals, or to analyze a signal or signals originating therefrom, and to provide an own voice beamformer weight control signal.
1 . A hearing device configured to be worn at or in an ear of the user, the hearing device comprising:
at least one first, outward-facing, input transducer configured to pick up first sounds from the environment of the user and to provide respective first electric input signals representative of said first sounds;
a second, inward-facing, input transducer configured to pick up a second sounds at the eardrum of the user and to provide a second electric input signal representative of said second sounds;
a directional system configured to receive said at least one first and said second electric input signals and comprising an own voice beamformer configured to provide an estimate of the user's own voice in dependence of said at least one first and said second electric input signals and configurable own voice beamformer weights;
an own voice analyzer configured to analyze at least one of said at least one first and said second electric input signals, or to analyze a signal or signals originating therefrom, and to provide an own voice beamformer weight control signal, the own voice analyzer comprising:
an own voice detector configured to estimate whether or not, or with what probability, a given input signal representative of sound originates from the voice of the user of the system, and to provide an own voice control signal indicative thereof;
an own voice classifier configured to classify, in dependence of the own voice control signal, an utterance from the user in at least two classes and to provide an own voice classification signal, and
an own voice beamformer controller configured to control said configurable own voice beamformer weights in dependence of said own voice classification signal, and providing said own voice beamformer weight control signal.
2 . A hearing device according to claim 1 , wherein the own voice beamformer controller is configured to determine or select a current steering vector in dependence of the own voice classification signal and to determine the configurable own voice beamformer weights in dependence thereon.
3 . A hearing device according to claim 1 , the hearing device comprising an ITE-part adapted to be located at or fully or partially in an ear canal of the user, wherein said ITE-part comprises said second, inward-facing, input transducer.
4 . A hearing device according to claim 1 , wherein said second, inward-facing, input transducer comprises a vibration sensor.
5 . A hearing device according to claim 1 , wherein said second, inward-facing, input transducer comprises an accelerometer.
6 . A hearing device according to claim 1 , wherein the own voice classifier is configured to provide the own voice classification signal in dependence of the content of the utterance.
7 . A hearing device according to claim 1 , wherein the own voice classifier is configured to provide the own voice classification signal in dependence of whether or not, or to which degree, a specific speech component is transmitted through the skull.
8 . A hearing device according to claim 1 , wherein the own voice classifier is configured to provide the own voice classification signal in dependence of one or more of: the energy of the signal, the degree of voicing, e.g., unvoiced or voiced, a fundamental frequency, e.g. pitch content, a spectral tilt, a spectral centroid, and a spectral flatness.
9 . A hearing device according to claim 1 , wherein the own voice classifier is configured to provide the own voice classification signal in dependence of a difference between the utterance picked up by the inward-facing microphone and the outward-facing microphone, e.g. a difference in input level or a difference in power spectral density of the corresponding electric input signals.
10 . A hearing device according to claim 1 , wherein the own voice classifier is based on a learning algorithm, e.g. a neural network, such as a deep neural network.
11 . A hearing device according to claim 1 , the hearing aid comprising a transform unit for converting a time domain signal to a signal in the transform domain.
12 . A hearing device according to claim 1 , wherein the hearing device is configured to provide that the configuration of the own voice beamformer weights vary across time and frequency.
13 . A hearing device according to claim 1 , the hearing device comprising a BTE-part adapted to be located at or behind an ear of said user, and wherein at least one of said at least one first, outward-facing, input transducers is/are located.
14 . A hearing device according to claim 1 , the hearing device comprising antenna and transmitter circuitry allowing the estimate of the user's own voice, or a processed version thereof, to be transmitted to another device or system.
15 . A hearing device according to claim 1 , the hearing device being constituted by or comprising an air-conduction type hearing aid or a headset or earphones, or a combination thereof.