Microphone probe, method, system and computer program product for audio signals processing

• US 10,455,323 B2
• Filed: 02/09/2017
• Issued: 10/22/2019
• Est. Priority Date: 02/09/2016
• Status: Active Grant

• Current Assignee: Zylia Spolka Z Ograniczona Odpowiedzialnoscia
• Original Assignee: Zylia Spolka Z Ograniczona Odpowiedzialnoscia
• Inventors: Januszkiewicz, Lukasz, Zernicki, Tomasz, Kurc, Maciej, Chryszczanowicz, Marcin, Zamojski, Jakub, Makaruk, Piotr, Szczechowiak, Piotr
• Primary Examiner(s): Sniezek, Andrew L
• Application Number:
  US16/076,951
• Publication Number:
  US 20190052957A1
• Time in Patent Office: 985 Days
• CPC Class Codes: H04R 1/406 microphones
• CPC Class Codes: H04R 19/04 Microphones H04R19/01 takes precedence
• CPC Class Codes: H04R 2201/003 Mems transducers or their use of the electrostatic type H04R19/005
• CPC Class Codes: H04R 2201/401 2D or 3D arrays of transducers
• CPC Class Codes: H04R 3/005 for combining the signals of two or more microphones specially adapted for hearing aids H04R25/407

Abstract

The invention concerns a microphone probe having a body being substantially a first solid of revolution with a number of audio sensors distributed thereon and located in the recesses. The recesses have substantially a shape of a second body of revolution with an axis of symmetry perpendicular to the surface of the body. The sensors are connected to an acquisition unit, that delivers audio signals to the output. The audio sensors are digital audio sensors comprising printed circuit board with MEMS microphone element mounted thereon, wherein MEMS microphone element is mounted on the side of the printed circuit board facing the inner side of the body, so that the sound reaches MEMS microphone element via recess and opening. The depth of recesses is in a range between 3 and 20 mm. The acquisition unit has a clocking device determining common time base for audio sensors.

Claims

1. A microphone probe having a first body being substantially a first solid of revolution with a number of audio sensors distributed thereon and located in recesses having substantially a shape of a second body of revolution having an axis of symmetry perpendicular to a surface of the first body, wherein the sensors are connected to an acquisition unit, characterized in that the audio sensors are digital audio sensors comprising a printed circuit board with at least one MEMS (microelectromechanical) microphone element mounted thereon, wherein the at least one MEMS microphone element is mounted on the side of the printed circuit board facing the interior of the first body, so that the sound reaches the at least one MEMS microphone element via the recess in the first body and an opening in the printed circuit board, wherein the depth of the recesses is in a range between 3 and 20 mm, and wherein the acquisition unit has a clocking device determining a common time base for the digital audio sensors, and wherein the acquisition unit is adapted to feed signals from particular digital audio sensors to a processing unit.

2. The microphone probe according to claim 1, characterized in that the processing unit is integrated with the microphone probe.

3. The microphone probe according to claim 1, characterized in that the acquisition unit is implemented as an FPGA (field-programmable gate array) unit with B_F-bit logic while the audio sensors provide B_S-bit samples, wherein B_F is lower or equal to B_S, and wherein a conversion is done with a module having a (2B_S−

B_F)-bit buffer, the module being adapted to;

write a sample into the buffer setting bits from 0 to (B_S−

1) with the bits of the sample and setting bits from B_S to (2B_S−

B_F−

1) with the value of the (B_S−

1)-th bit of the sample;

apply a gain by shifting the bits of the buffer to the left by a given number of positions;

detect saturation when either bit number (2B_S−

B_F−

1) is “

0” and

bits from (2B_S−

B_F−

2) to (B_S−

1) are filed with “

1”

or bit number (2B_S−

B_F−

1) is “

1” and

bits from (2B_S−

B_F−

2) to (B_S−

1) are filed with “

0”

; and

return either saturation information or the value of the bits from (B_S−

1) to (B_S−

B_F) of the buffer as a return value.

4. The microphone probe according to claim 3, characterized in that B_F is equal to 16 and B_S is equal to 24.

5. The microphone probe according to claim 1, characterized in that the first body is substantially spherical.

6. The microphone probe according to claim 5, characterized in that digital audio sensors are distributed in evenly spaced layers.

7. The microphone probe according to claim 5, characterized in that the digital audio sensors are distributed in parallel layers corresponding to evenly distributed angles of latitude.

8. The microphone probe according to claim 5, characterized in that it has at least 19 audio sensors.

9. The microphone probe according to claim 8 characterized in that it has 62 audio sensors.

10. The microphone probe according to claim 1, characterized in that the first body is substantially cylindrical and the audio sensors are uniformly distributed on its lateral surface.

11. A method of processing audio signals comprising the steps of:

acquiring a number of N signals from audio sensors;

determining a direction of arrival of sound originating from a number of M sources;

applying beamforming to obtain M channels corresponding to these sources from acquired signals using a filter table,

characterized in that the frequency band of the acquired signals is divided at least into a first frequency band and a second frequency band, while a first beamforming method is applied in the first frequency band and a second beamforming method is applied in the second frequency band; and

applying postprocessing including filtration of at least one of the M channels with a source-specific filtration wherein the value of the number of audio sensors used in beamforming depends on the frequency band and is selected so that the spacing between sensors is greater than 0.05 of the wavelength and lower than 0.5 of the wavelength in each of the frequency bands.

12. The method according to claim 11, characterized in that determining the direction of arrival of the sound originating from the M sources includes receiving at least partial indication of the location of at least one source with user interface prior, during, or after the acquisition.

13. The method according to claim 12, characterized in that the reception of at least partial indication of the location of at least one source with user interfaces precedes the acquisition of the N signals from the audio sensors and in that an additional step of determining the impulse response or the transmittance of a link between at least one source and the audio sensors is executed before the acquisition, wherein the measured impulse response or the transmittance is used to compensate the effect of environment on the sound from at least one source.

14. The method according to claim 11, characterized in that filtration includes adaptive Wiener filtration of at least first channel including adaptive filtering and subtraction of signals from at least two other channels.

15. The method according to claim 11, characterized in that the beamforming is based on correlation matrix between signals of the audio sensors.

16. The method according to claim 11, characterized in that the beamforming is based on a frequency response matrix of the audio sensors.

17. The method according to claim 16, characterized in that the frequency response matrix of the audio sensors is a result of the prior measurements in an anechoic chamber.

18. An audio acquisition system comprising a microphone probe, a processing unit, and an external interface, the microphone probe

having a first body being substantially a first solid of revolution with a number of audio sensors distributed thereon and located in recesses having substantially a shape of a second body of revolution having an axis of symmetry perpendicular to a surface of the first body,

wherein the sensors are connected to an acquisition unit, characterized in that the audio sensors are digital audio sensors comprising a printed circuit board with at least one MEMS (microelectromechanical) microphone element mounted thereon, wherein the at least one MEMS microphone element is mounted on the side of the printed circuit board facing the interior of the first body, so that the sound reaches the at least one MEMS microphone element via the recess in the first body and an opening in the printed circuit board, wherein the depth of the recesses is in a range between 3 and 20 mm, and wherein the acquisition unit has a clocking device determining a common time base for the digital audio sensors, and wherein the acquisition unit is adapted to feed the signals from particular digital audio sensors to, a processing unit which is adapted to carry on a method comprising the steps of:

acquiring a number N of signals from audio sensors;

determining a direction of arrival of sound originating from a M number of sources;

applying beamforming to obtain M channels corresponding to these sources from acquired signals using a filter table;

characterized in that the frequency band of the acquired signals is divided at least into a first frequency band and a second frequency band, while a first beamforming method is applied in the first frequency band and a second beamforming method is applied in the second frequency band;

applying postprocessing including filtration of at least one of the M channels with a source-specific filtration; and

outputting resulting channels with the external interface.