Method at a binaural hearing device system and a binaural hearing device system
Disclosed is a method at a binaural hearing device system, the system comprises a first hearing device for placement at, or in, a user's first ear, the first hearing device comprising a first acoustic input transducer arrangement, a first signal processor, and a first wireless communication interface; and a second hearing device for placement at, or in, the user's second ear, the second hearing device comprising a second acoustic input transducer arrangement, a second signal processor, and a second wireless communication interface. The method comprises, at a first signal processor, computing a first bilateral beamforming signal based on first and second monaural beamforming signals. Computing the first bilateral beamforming signal comprises an adaptive computation including a variable length FIR filter and a coefficient-based adaptation strategy.
1 . A method performed by a binaural hearing device system, the system comprising (1) a first hearing device for placement at, or in, a first ear of a user, the first hearing device comprising a first input transducer arrangement, a first signal processor, and a first wireless communication interface, and (2) a second hearing device for placement at, or in, a second ear of the user, the second hearing device comprising a second input transducer arrangement, a second signal processor, and a second wireless communication interface, the method comprising:
generating a first monaural beamforming signal by the first signal processor based on one or more input transducer signals provided by the first input transducer arrangement;
obtaining a second monaural beamforming signal provided by the second hearing device; and
determining a first bilateral beamforming signal based on the first and second monaural beamforming signals, wherein the first bilateral beamforming signal is determined based on a variable length finite impulse response (FIR) filter, the FIR filter having a filter length;
wherein the first bilateral beamforming signal is based on α*l+β*r, wherein l is one of the first and second monaural beamforming signals, r is the other one of the first and second monaural beamforming signals, * represents a convolution operation, and α and β relate to the variable length FIR filter.
2 . The method of claim 1 , wherein the first bilateral beamforming signal is determined based on a first value representing a number of one or more filter coefficients to be adapted per signal block.
3 . The method of claim 1 , further comprising receiving a request for an optimization mode.
4 . The method of claim 3 , wherein the request for the optimization mode is based on a user input.
5 . The method of claim 3 , wherein the request for the optimization mode is based on a listening environment.
6 . The method of claim 3 , wherein the request is automatically generated.
7 . The method of claim 3 , wherein the filter length comprises a first filter length, and wherein the method further comprises determining a second filter length based on a current listening environment when in the optimization mode.
8 . The method of claim 1 , wherein when the binaural hearing device system is in a first mode, the filter length has a first filter length value, and/or a number of one or more filter coefficients to be adapted per signal block has a first value.
9 . The method of claim 8 , wherein the first filter length value has a predetermined default filter length value, and the first value of the number of the one or more filter coefficients to be adapted per signal block is a predetermined default value.
10 . The method of claim 8 , wherein when the binaural hearing device system is in a second mode, the filter length has a second filter length value, and/or wherein the number of the one or more filter coefficients to be adapted per signal block has a second value.
11 . The method of claim 10 , wherein the first mode and/or the second mode is based on a user input.
12 . The method of claim 10 , wherein the first mode and/or the second mode is based on a listening environment.
13 . The method of claim 1 , wherein a number of one or more filter coefficients to be adapted per signal block has a first value when the binaural hearing device system is in a first mode, and the number of the one or more filter coefficients to be adapted per signal block has a second value when the binaural hearing device system is in a second mode.
14 . The method of claim 13 , wherein the second value of the number of the one or more filter coefficients to be adapted per signal block is determined by machine learning.
15 . The method of claim 1 , further comprising determining a listening environment using machine learning.
16 . The method of claim 1 , wherein the determining the first bilateral beamforming signal comprises performing a time-domain adaptive computation.
17 . The method of claim 1 , wherein a number of the one or more filter coefficients to be adapted per signal block is same as, or is less than, a total number of all filter coefficients of the variable length FIR filter.
18 . The method of claim 17 , wherein the one or more filter coefficients are a subset of all filter coefficients of the variable length FIR filter, and wherein the method further comprises setting remaining coefficients that are not in the subset to a default value.
19 . The method of claim 1 , further comprising segmenting the first and second monaural beamforming signals into a number of signal blocks, wherein the number of the signal blocks is anywhere from 16 to 96.
20 . The method of claim 1 , further comprising updating a number of one or more filter coefficients to be adapted per signal block, wherein the act of updating is performed using a counter.
21 . The method of claim 1 , further comprising slowing down an adaptation convergence when there is a change of listening environment.
22 . The method of claim 1 , wherein the variable length FIR filter is configured to equalize a head-shadow effect between the first monaural beamforming signal and the second monaural beamforming signal.
23 . The method of claim 1 , wherein the variable length FIR filter is configured to equalize a beamforming difference between the first monaural beamforming signal and the second monaural beamforming signal.
24 . The method of claim 1 , further comprising adjusting a speed of convergence for computation of the first bilateral beamforming signal.
25 . The method of claim 24 , wherein the act of adjusting the speed of convergence comprises reducing the speed of convergence.
26 . The method of claim 1 , wherein α+β=1 in a first situation, and α+β=0 in a second situation.
27 . A binaural hearing device system, comprising:
a first hearing device for placement at, or in, a first ear of a user, the first hearing device comprising a first input transducer arrangement, a first signal processor, and a first wireless communication interface; and
a second hearing device for placement at, or in, a second ear of the user, the second hearing device comprising a second input transducer arrangement, a second signal processor, and a second wireless communication interface; and
wherein the first signal processor is configured to:
generate a first monaural beamforming signal based on one or more input transducer signals provided by the first input transducer arrangement;
obtain a second monaural beamforming signal provided by the second hearing device; and
determine a first bilateral beamforming signal based on the first and second monaural beamforming signals, wherein the first bilateral beamforming signal is determined based on a variable length finite impulse response (FIR) filter, the FIR filter having a filter length;
wherein the first bilateral beamforming signal is based on α*l+β*r, wherein l is one of the first and second monaural beamforming signals, r is the other one of the first and second monaural beamforming signals, * represents a convolution operation, and α and β relate to the variable length FIR filter.
28 . The binaural hearing device system of claim 27 , wherein the first signal processor is configured to slow down an adaptation convergence when there is a change of listening environment.
29 . The binaural hearing device system of claim 27 , wherein the variable length FIR filter is configured to equalize a head-shadow effect between the first monaural beamforming signal and the second monaural beamforming signal.
30 . The binaural hearing device system of claim 27 , wherein the variable length FIR filter is configured to equalize a beamforming difference between the first monaural beamforming signal and the second monaural beamforming signal.
31 . The binaural hearing device system of claim 27 , wherein the binaural hearing device system is configured to adjust a speed of convergence for computation of the first bilateral beamforming signal.
32 . The binaural hearing device system of claim 31 , wherein the binaural hearing device system is configured to adjust the speed of convergence by reducing the speed of convergence.
33 . The binaural hearing device system of claim 27 , wherein α+β=1 in a first situation, and α+β=0 in a second situation.