# US 2021/0084427 A1 · Audio Diarization System that Segments Audio Input

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

- Number: US 2021/0084427 A1
- Document ID: US20210084427A1
- Status: Published application
- Kind: A1
- Application number: 17/092,200
- Filed: 2020-11-06
- Published: 2021-03-18
- Earliest claimed: 2016-06-10
- Inventors: Philip Scott Lyren; Glen A. Norris
- Web page: https://patentgrep.com/patent/US20210084427A1

## CPC

- H04S 7/30: {Control circuits for electronic adaptation of the sound field}
- G10L 15/04: Segmentation; Word boundary detection
- G10L 25/81: for discriminating voice from music
- H02M 1/08: Circuits specially adapted for the generation of control voltages for semiconductor devices incorporated in static converters
- H02M 1/4208: {Arrangements for improving power factor of AC input}
- H04M 1/0202: {Portable telephone sets, e.g. cordless phones, mobile phones or bar type handsets (constructional features of telephone transmitters or receivers, e.g. of speakers or microphones H04M1/03)}
- H04W 88/02: Terminal devices
- H02M 3/24: by static converters
- H04M 2201/40: using speech recognition
- H04S 7/302: {Electronic adaptation of stereophonic sound system to listener position or orientation (H04S7/301 takes precedence)}
- H04S 2420/01: Enhancing the perception of the sound image or of the spatial distribution using head related transfer functions [HRTF's] or equivalents thereof, e.g. interaural time difference [ITD] or interaural level difference [ILD]
- H04S: STEREOPHONIC SYSTEMS
- G10L: SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- H02M: APPARATUS FOR CONVERSION BETWEEN AC AND AC, BETWEEN AC AND DC, OR BETWEEN DC AND DC, AND FOR USE WITH MAINS OR SIMILAR POWER SUPPLY SYSTEMS; CONVERSION OF DC OR AC INPUT POWER INTO SURGE OUTPUT POWER; CONTROL OR REGULATION THEREOF (transformers H01F; dynamo-electric converters H02K47/00; controlling transformers, reactors or choke coils, control or regulation of electric motors, generators or dynamo-electric converters H02P)
- H04M: TELEPHONIC COMMUNICATION (circuits for controlling other apparatus via a telephone cable and not involving telephone switching apparatus G08)
- H04W: WIRELESS COMMUNICATION NETWORKS (broadcast communication H04H; communication systems using wireless links for non-selective communication, e.g. wireless extensions H04M1/72)

## Abstract

A wearable electronic device worn on a head of a user analyzes prior electronic communications and determines where to play a voice so the voice originates at a familiar sound localization point (SLP) to the user. The wearable electronic device plays the voice so the voice externally localizes to the user as binaural sound at the familiar SLP.

## Claims

### Claim at position 1

-20. (canceled)

### Claim 21

A method comprising:
determining, by a wearable electronic device (WED) worn on a head of a person, a familiar sound localization point (SLP) by analyzing SLPs in prior electronic communications where voices in the prior electronic communications externally localized as binaural sound to the person;
processing, with one or more processors, a voice in an electronic communication so the voice externally localizes as binaural sound from the familiar SLP that is external to the person and at a far field distance from the person; and
providing, by the WED and with earphones or headphones, the voice in the binaural sound to the person so the voice externally localizes from the familiar SLP that is external to the person and at the far field distance from the person.

### Claim 22 (depends on 21)

The method of claim 21, further comprising:
executing speech recognition to identify the voice in the electronic communication; and
determining, from the speech recognition, that the familiar SLP is assigned to the voice identified from the speech recognition.

### Claim 23 (depends on 21)

The method of claim 21, further comprising:
receiving, during the electronic communication that includes multiple different voices that internally localize to the person, a command to move one of the voices to externally localize while other ones of the multiple different voices internally localized.

### Claim 24 (depends on 21)

The method of claim 21, further comprising:
segmenting sounds during the electronic communication so one of the sounds internally localizes inside the head of the person as mono or stereo sound and another one of the sounds externally localizes to the person as the binaural sound to the familiar SLP.

### Claim 25 (depends on 21)

The method of claim 21, further comprising:
prefetching, before the electronic communication commences, head-related transfer functions (HRTFs) in order to increase execution speed of processing the voice in the electronic communication.

### Claim 26 (depends on 21)

The method of claim 21, further comprising:
assigning telephone numbers to the SLPs;
storing, in memory of the WED, the telephone numbers and the SLPs; and
retrieving, from the memory, a SLP for a telephone number of a caller upon identifying the telephone number.

### Claim 27 (depends on 21)

The method of claim 21 further comprising:
caching, in the WED and before the electronic communication commences, head-related transfer functions (HRTFs) in order to increase execution speed of processing the voice in the electronic communication.

### Claim 28

A method to process a voice in an electronic communication to an external sound localization point (SLP) that is familiar to a user, the method comprising:
receiving, with a wearable electronic device (WED) worn on a head of the user, the electronic communication that will externally localize to the user as binaural sound;
determining, based on analysis of prior electronic communications, a preferred SLP that is familiar to the user where the user previously localized voices in prior electronic communications; and
processing, with one or more processors, the voice in the electronic communication so the voice plays through earphones or headphones and externally localizes to the user at the preferred SLP that is familiar to the user, wherein the preferred SLP is at a far field distance from the user.

### Claim 29 (depends on 28)

The method according to claim 28 further comprising:
storing, in memory of the WED, the preferred SLP where the user desires to localize the voice in the electronic communication.

### Claim 30 (depends on 28)

The method according to claim 28 further comprising:
prefetching, by the WED, head-related transfer functions (HRTFs) that are processed with the voice in order to increase execution speed of processing the voice in the electronic communication into the binaural sound.

### Claim 31 (depends on 28)

The method according to claim 28 further comprising:
receiving a telephone call from an unknown telephone number;
retrieving an SLP assigned to unknown telephone numbers; and
processing sound from the unknown telephone number to localize to the SLP assigned to the unknown telephone numbers.

### Claim 32 (depends on 28)

The method according to claim 28 further comprising:
analyzing historic information stored in memory to determine where to localize the voice in the electronic communication, wherein the historic information includes coordinate locations of head-related transfer functions (HRTFs) processed in the prior electronic communications.

### Claim 33 (depends on 28)

The method according to claim 28 further comprising:
storing, in memory, user preferences of the user that include locations in spherical coordinates where the user previously externally localized the voices in the prior electronic communications.

### Claim 34 (depends on 28)

The method according to claim 28 further comprising:
assigning head-related transfer functions (HRTFs) to telephone numbers;
determining a telephone number of an incoming electronic communication;
selecting a pair of the HRTFs that are assigned to the telephone number of the incoming electronic communication; and
processing the incoming electronic communication with the pair of the HRTFs that are assigned to the telephone number of the incoming electronic communication.

### Claim 35 (depends on 28)

The method according to claim 28 further comprising:
caching, in memory of the WED, head-related transfer functions (HRTFs) that process the voice in order to increase execution speed of processing the voice in the electronic communication.

### Claim 36

A wearable electronic device (WED) worn on a head of a user, the WED comprising:
one or more processors that execute instructions that determine a familiar sound localization point (SLP) where voices in prior electronic communications externally localized as binaural sound to the user and that process a voice in an electronic communication so the voice externally localizes as binaural sound from the familiar SLP that is external to the user at a far field distance from the user; and
speakers that play the voice in the binaural sound to the user so the voice externally localizes from the familiar SLP that is external to the user and at the far field distance from the user.

### Claim 37 (depends on 36)

The WED of claim 36 further comprising:
a memory that stores a table that includes assignments of SLPs to telephone numbers.

### Claim 38 (depends on 36)

The WED of claim 36, wherein the one or more processors expedite processing of the voice by prefetching head-related transfer functions (HRTFs), wherein the prefetching occurs after the WED receives the electronic communication but before the user answers the electronic communication.

### Claim 39 (depends on 36)

The WED of claim 36, wherein the one or more processors further execute the instructions to predict where the user prefers to localize the voice in the electronic communication based on previous SLPs where the user already localized voices in the prior electronic communications.

### Claim 40 (depends on 36)

The WED of claim 36, wherein the one or more processors include a digital signal processor (DSP) with a specialized hardware architecture to increase execution speed of processing sound for the electronic communication by prefetching and caching head-related transfer functions (HRTFs).
