# US 12,015,908 B2 · Method and apparatus for recognizing sound source

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 12,015,908 B2
- Document ID: US12015908B2
- Status: Granted
- Kind: B2
- Application number: 17/424,128
- Filed: 2019-11-14
- Published: 2022-04-28
- Granted: 2024-06-18
- Earliest claimed: 2019-01-28
- Assignees: Young Eon Kim
- Applicants: Young Eon Kim
- Inventors: Young Eon Kim; Min Soo Kim
- Web page: https://patentgrep.com/patent/US12015908B2

## CPC

- H04S 7/302: {Electronic adaptation of stereophonic sound system to listener position or orientation (H04S7/301 takes precedence)}
- G01S 11/14: using ultrasonic, sonic or infrasonic waves
- G08G 1/0965: responding to signals from another vehicle, e.g. emergency vehicle
- G10L 21/0272: Voice signal separating
- H04R 3/005: {for combining the signals of two or more microphones (specially adapted for hearing aids H04R25/407)}
- H04R 3/04: for correcting frequency response
- H04R 5/027: Spatial or constructional arrangements of microphones, e.g. in dummy heads
- H04R 5/04: Circuit arrangements, {e.g. for selective connection of amplifier inputs/outputs to loudspeakers, for loudspeaker detection, or for adaptation of settings to personal preferences or hearing impairments (combinations of amplifiers H03F3/68; stereophonic systems H04S)}
- H04S 3/002: {Non-adaptive circuits, e.g. manually adjustable or static, for enhancing the sound image or the spatial distribution (control circuits for electronic adaptation of the sound field H04S7/30)}
- H04S 7/307: {Frequency adjustment, e.g. tone control (H04S7/301 takes precedence)}
- H04S: STEREOPHONIC SYSTEMS
- G01S: RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G08G: TRAFFIC CONTROL SYSTEMS (guiding railway traffic, ensuring the safety of railway traffic B61L; arrangement of road signs or traffic signals E01F9/00; radar or analogous systems, sonar systems, lidar systems specially adapted for traffic control G01S13/91, G01S15/88, G01S17/88; {radar or analogous systems, sonar systems, lidar systems specially adapted for anti-collision purposes G01S13/93, G01S15/93, G01S17/93})
- G10L: SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- H04R: LOUDSPEAKERS, MICROPHONES, GRAMOPHONE PICK-UPS OR LIKE ACOUSTIC ELECTROMECHANICAL TRANSDUCERS; ELECTRIC HEARING AIDS; PUBLIC ADDRESS SYSTEMS (producing sounds with frequency not determined by supply frequency G10K)

## Abstract

The present invention relates to a method and apparatus for recognizing a sound source. According to the sound source recognition method of the present invention, acoustic signals are detected from four acoustic sensors arranged in a rectangular shape when viewed in a horizontal direction, sound arrival times are measured, six Interactive Time Differences (ITDs) are generated based on differences in sound arrival time between the respective acoustic sensors, and the location of a sound source is estimated based on the six ITDs. In addition, the type of sound source is determined by extracting and classifying features of the sound source by using a sum signal from the four acoustic sensors.

## Claims

### Claim 1

A method for recognizing a sound source, the method comprising:
detecting acoustic signals using at least four acoustic sensors, each of the acoustic sensors detecting the acoustic signals from air, wherein four of the acoustic sensors are arranged at vertices A, B, C, and D of a specific rectangle in a plan view, respectively, and acoustic signals detected by the four acoustic sensors are referred to as A(s), B(s), C(s), and D(s), respectively;
calculating sound arrival times from the respective acoustic signals A(s), B(s), C(s), and D(s);
generating six interaural time differences (ITDs) based on a difference in sound arrival time between the acoustic sensor at vertex A and the acoustic sensor at vertex B, between the acoustic sensor at vertex B and the acoustic sensor at vertex C, between the acoustic sensor at vertex C and the acoustic sensor at vertex D, between the acoustic sensor at vertex D and the acoustic sensor at vertex A, between the acoustic sensor at vertex A and the acoustic sensor at vertex C, and between the acoustic sensor at vertex B and the acoustic sensor at vertex D; and
estimating a location of a sound source based on at least two of the six ITDs,
wherein estimating the location of the sound source comprises:
calculating an azimuth angle θ1 formed by a line connecting a first pair of acoustic sensors at the vertices A and D of the four acoustic sensors and a line connecting the sound source and the acoustic sensor at the vertex D based on an ITD between the first pair of acoustic sensors at the vertices A and D;
calculating an azimuth angle θ2 formed by a line connecting a second pair of acoustic sensors at the vertices B and A of the four acoustic sensors and a line connecting the sound source and the acoustic sensor at the vertex A based on an ITD between the second pair of acoustic sensors at the vertices B and A, wherein the first pair of acoustic sensors at the vertices A and D and the second pair of acoustic sensors at the vertices B and A share the acoustic sensor at the vertex A; and
calculating a distance to the sound source using the calculated azimuth angles θ1 and θ2, a distance VL between the first pair of acoustic sensors at the vertices A and D, and a distance VW between the second pair of acoustic sensors at the vertices B and A.

### Claim 2 (depends on 1)

The method of claim 1, further comprising:
calculating at least one of signals y(s), f(s), b(s), I(s), r(s), cl(s), cr(s), p(s), or q(s) by combining the acoustic signals A(s), B(s), C(s), or D(s) detected by the four acoustic sensors, as follows:
y(s)=A(s)+B(s)+C(s)+D(s);
f(s)=A(s)+B(s);
b(s)=C(s)+D(s);
l(s)=A(s)+D(s);
r(s)=B(s)+C(s);
cl(s)=A(s)+C(s);
cr(s)=B(s)+D(s);
p(s)=f(s)−b(s); and
q(s)=l(s)−r(s),
estimating at least one of a volume level, direction, and moving direction of the sound source based on the signals y(s), f(s), b(s), I(s), r(s), cl(s), cr(s), p(s), or q(s).

### Claim 3 (depends on 1)

The method of claim 1, further comprising determining a type of sound source by extracting a feature of the sound source using a signal y(s), which is a sum signal of the four sound signals, and classifying the extracted feature.

### Claim 4 (depends on 1)

The method of claim 1, wherein calculating the distance to the sound source further comprises correcting an error by adopting an error correction function in order to correct the error occurring in the calculation of the distance to the sound source.

### Claim 5 (depends on 1)

The method of claim 1, further comprising, before detecting the acoustic signals, performing trimming so that same output signals are output from the four acoustic sensors in a state without an input signal in order to perform initialization.

### Claim 6 (depends on 1)

The method of claim 1, wherein detecting the acoustic signals comprises:
canceling voice signals from signals input to the four acoustic sensors.

### Claim 7 (depends on 6)

The method of claim 6, wherein detecting the acoustic signals comprises:
removing noise signals common to the four acoustic sensors from the signals of the four acoustic sensors from which the voice signals have been canceled, and outputting resulting acoustic signals.

### Claim 8 (depends on 1)

The method of claim 1, wherein at least one of the four acoustic sensors is disposed at a height different from that of remaining acoustic sensors.

### Claim 9 (depends on 1)

The method of claim 1, wherein the acoustic sensors comprise noise shielding acoustic sensors around each of which a shield block is installed to suppress unnecessary signals.

### Claim 10

An apparatus for recognizing a sound source, the apparatus comprising:
at least four acoustic sensors configured to detect acoustic signals, each of the acoustic sensors detecting the acoustic signals from air, wherein four of the acoustic sensors are arranged at vertices A, B, C, and D of a specific rectangle in a plan view, respectively, and acoustic signals detected by the four acoustic sensors are referred to as A(s), B(s), C(s), and D(s), respectively;
a sound arrival time measurement unit configured to calculate sound arrival times from the respective acoustic signals A(s), B(s), C(s), and D(s);
an ITD generation unit configured to generate six interaural time differences (ITDs) based on a difference in sound arrival time between the acoustic sensor at vertex A and the acoustic sensor at vertex B, between the acoustic sensor at vertex B and the acoustic sensor at vertex C, between the acoustic sensor at vertex C and the acoustic sensor at vertex D, between the acoustic sensor at vertex D and the acoustic sensor at vertex A, between the acoustic sensor at vertex A and the acoustic sensor at vertex C, and between the acoustic sensor at vertex B and the acoustic sensor at vertex D; and
a sound source location estimation unit configured to estimate a location of a sound source based on at least two of the six ITDs,
wherein estimating the location of the sound source comprises:
calculating an azimuth angle θ1 formed by a line connecting a first pair of acoustic sensors at the vertices A and D of the four acoustic sensors and a line connecting the sound source and the acoustic sensor at the vertex D based on an ITD between the first pair of acoustic sensors at the vertices A and D;
calculating an azimuth angle θ2 formed by a line connecting a second pair of acoustic sensors at the vertices B and A of the four acoustic sensors and a line connecting the sound source and the acoustic sensor at the vertex A based on an ITD between the second pair of acoustic sensors at the vertices B and A, wherein the first pair of acoustic sensors at the vertices A and D and the second pair of acoustic sensors at the vertices B and A share the acoustic sensor at the vertex A; and
calculating a distance to the sound source using the calculated azimuth angles θ1 and θ2, a distance VL between the first pair of acoustic sensors at the vertices A and D, and a distance VW between the second pair of acoustic sensors at the vertices B and A.

### Claim 11 (depends on 1)

A non-transitory computer-readable medium storing codes for causing a computer to perform the method of claim 1.