# US 2022/0068298 A1 · Wearable mixed sensor array for self-voice capture

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

- Number: US 2022/0068298 A1
- Document ID: US20220068298A1
- Status: Published application
- Kind: A1
- Application number: 17/002,254
- Filed: 2020-08-25
- Published: 2022-03-03
- Earliest claimed: 2020-08-25
- Assignees: Bose Corporation
- Applicants: Bose Corporation
- Inventors: Aric J. Wax; Michael Shay
- Web page: https://patentgrep.com/patent/US20220068298A1

## CPC

- G10L 25/84: for discriminating voice from noise
- G06F 3/011: {Arrangements for interaction with the human body, e.g. for user immersion in virtual reality (blind teaching G09B21/00)}
- G06F 3/017: {Gesture based interaction, e.g. based on a set of recognized hand gestures (interaction based on gestures traced on a digitiser G06F3/04883)}
- G06N 20/00: Machine learning
- H04R 3/005: {for combining the signals of two or more microphones (specially adapted for hearing aids H04R25/407)}
- H04R 5/033: Headphones for stereophonic communication {(details thereof, e.g. relating to batteries, cables or control elements H04R1/10)}
- 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)}
- H04R 1/406: {microphones}
- H04R 2201/107: Monophonic and stereophonic headphones with microphone for two-way hands free communication
- H04R 2410/05: Noise reduction with a separate noise microphone
- H04R 2430/20: Processing of the output signals of the acoustic transducers of an array for obtaining a desired directivity characteristic (H04R2203/12 takes precedence)
- H04R 2460/13: Hearing devices using bone conduction transducers
- G10L: SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- 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

A wearable for capturing self-voice using a mixed sensor array, including a microphone generating a microphone signal and positioned to capture a user's voice; an accelerometer generating an accelerometer signal and positioned to capture the user's voice; and a controller receiving the microphone signal and the accelerometer signal and configured to generate a mixed output signal, wherein the mixed output signal comprises a microphone signal component and an accelerometer signal component, wherein a ratio of the microphone signal component and the accelerometer signal component is related to a value of a noise metric representing a magnitude of ambient noise.

## Claims

### Claim 1

A wearable for capturing self-voice using a mixed sensor array, comprising:
a microphone generating a microphone signal and positioned to capture a user's voice;
an accelerometer generating an accelerometer signal and positioned to capture the user's voice; and
a controller receiving the microphone signal and the accelerometer signal and configured to generate a mixed output signal, wherein the mixed output signal comprises a microphone signal component and an accelerometer signal component, wherein a ratio of the microphone signal component and the accelerometer signal component is related to a value of a noise metric representing a magnitude of ambient noise.

### Claim 2 (depends on 1)

The wearable of claim 1, further comprising a second microphone generating a second microphone signal, wherein the microphone signal component is comprised of the microphone signal and the second microphone signal.

### Claim 3 (depends on 1)

The wearable of claim 1, further comprising a second accelerometer generating a second accelerometer signal, wherein the accelerometer signal component is comprised of the accelerometer signal and the second accelerometer signal.

### Claim 4 (depends on 1)

The wearable of claim 1, wherein the noise metric is determined according to a time-average of the microphone signal.

### Claim 5 (depends on 4)

The wearable of claim 4, wherein the time average of the microphone signal is calculated when a voice activity detector does not detect the user's voice.

### Claim 6 (depends on 4)

The wearable of claim 4, wherein the time average is spectrally weighted.

### Claim 7 (depends on 1)

The wearable of claim 1, wherein the noise metric is determined according to a machine learning algorithm.

### Claim 8 (depends on 2)

The wearable of claim 2, wherein the noise metric is determined according to a magnitude of a beam formed from, at least, the first microphone signal and the second microphone signal, the beam being directed to exclude the user's voice.

### Claim 9 (depends on 1)

The wearable of claim 1, wherein the ratio of the microphone signal component and the accelerometer signal component is determined according weights stored in a look up table.

### Claim 10 (depends on 1)

The wearable of claim 1, wherein the wearable comprises one of: headphones, earbuds, glasses, or a mobile device.

### Claim 11

A method for capturing self-voice with a mixed sensor array, comprising:
receiving a microphone signal from a microphone positioned to capture a user's voice;
receiving an accelerometer signal from an accelerometer positioned to capture the user's voice; and
generating a mixed output signal comprising a microphone signal component and an accelerometer signal component, wherein a ratio of the microphone signal component and the accelerometer signal component is related to a value of a noise metric representing a magnitude of ambient noise.

### Claim 12 (depends on 11)

The method of claim 11, further comprising the step of receiving a second microphone signal from a second microphone, wherein the microphone signal component is comprised of the microphone signal and the second microphone signal.

### Claim 13 (depends on 11)

The method of claim 11, further comprising the step of receiving a second accelerometer signal from a second accelerometer, wherein the accelerometer signal component is comprised of the accelerometer signal and the second accelerometer signal.

### Claim 14 (depends on 11)

The method of claim 11, wherein the noise metric is determined according to a time-average of the microphone signal.

### Claim 15 (depends on 14)

The method of claim 14, wherein the time average of the microphone signal is calculated when a voice activity detector does not detect the user's voice.

### Claim 16 (depends on 14)

The method of claim 14, wherein the time average is spectrally weighted.

### Claim 17 (depends on 11)

The method of claim 11, wherein the noise metric is determined according to a machine learning algorithm.

### Claim 18 (depends on 12)

The method of claim 12, wherein the noise metric is determined according to a magnitude of a beam formed from, at least, the first microphone signal and the second microphone signal, the beam being directed to exclude the user's voice.

### Claim 19 (depends on 11)

The method of claim 11, wherein the ratio of the microphone signal component and the accelerometer signal component is determined according weights stored in a look up table.

### Claim 20 (depends on 11)

The method of claim 11, wherein the wearable comprises one of: headphones, earbuds, glasses, or a mobile device.
