# US 2022/0044680 A1 · Privacy mode based on speaker identifier

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

- Number: US 2022/0044680 A1
- Document ID: US20220044680A1
- Status: Published application
- Kind: A1
- Application number: 17/463,965
- Filed: 2021-09-01
- Published: 2022-02-10
- Earliest claimed: 2017-06-02
- Assignees: Amazon Technologies, Inc.
- Applicants: Amazon Technologies, Inc.
- Inventors: Zhenhua Wang
- Web page: https://patentgrep.com/patent/US20220044680A1

## CPC

- G10L 15/22: Procedures used during a speech recognition process, e.g. man-machine dialogue
- G06F 3/167: {Audio in a user interface, e.g. using voice commands for navigating, audio feedback}
- G06F 21/32: using biometric data, e.g. fingerprints, iris scans or voiceprints
- G06F 21/6245: {Protecting personal data, e.g. for financial or medical purposes}
- G06F 21/64: Protecting data integrity, e.g. using checksums, certificates or signatures
- G10L 15/26: Speech to text systems (G10L15/08 takes precedence)
- G10L 17/00: Speaker identification or verification techniques
- G10L 2015/223: {Execution procedure of a spoken command}
- 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)

## Abstract

Techniques for configuring a speech processing system with a privacy mode that is associated with the identity of a user that activated the privacy mode are described. A user may speak an indication to have the speech processing system activate a privacy mode. When such an indication is detected by the speech processing system, the speech processing system determines an identity of the user, determines a unique system identifier associated with the user, and generates a privacy mode flag. The speech processing system then associates the privacy mode flag with the user's unique system identifier. The privacy mode flag indicates to components of the speech processing system that any data related to processing of the user's utterances should not be sent to long term storage, thus causing various components of the system to delete data once the respective component is finished processing with respect to an utterance of the user.

## Claims

### Claim at position 1

-20. (canceled)

### Claim 21

A computer-implemented method, comprising:
receiving first audio data corresponding to a first utterance, the first audio data corresponding to a first identifier;
performing first speech processing on the first audio data to determine results data;
receiving second audio data corresponding to a second utterance, the second audio data corresponding to the first identifier;
performing second speech processing on the second audio data to determine that the second utterance requests deletion of data corresponding to the first utterance;
based at least in part on the first identifier, identifying stored data corresponding to the first utterance; and
deleting the stored data.

### Claim 22 (depends on 21)

The computer-implemented method of claim 21, wherein the second utterance directly follows the first utterance with respect to the first identifier and wherein the second speech processing determines the second utterance refers to a previous utterance.

### Claim 23 (depends on 21)

The computer-implemented method of claim 21, wherein the stored data comprises at least a portion of the first audio data.

### Claim 24 (depends on 21)

The computer-implemented method of claim 21, wherein the stored data comprises a textual representation of the first utterance.

### Claim 25 (depends on 21)

The computer-implemented method of claim 21, wherein the stored data comprises first data linking the first utterance to usage history data.

### Claim 26 (depends on 21)

The computer-implemented method of claim 21, wherein the stored data comprises the results data.

### Claim 27 (depends on 21)

The computer-implemented method of claim 21, wherein the first identifier corresponds to a user identifier.

### Claim 28 (depends on 21)

The computer-implemented method of claim 21, wherein the first identifier corresponds to a session identifier.

### Claim 29 (depends on 21)

The computer-implemented method of claim 21, further comprising:
causing output of the results data,
wherein the second audio data is received after causing output of the results data.

### Claim 30 (depends on 21)

The computer-implemented method of claim 21, further comprising, prior to deleting the stored data:
determining the second utterance was spoken by a same user as the first utterance.

### Claim 31

A system comprising:
at least one processor; and
at least one memory comprising instructions that, when executed by the at least one processor, cause the system to:
receive first audio data corresponding to a first utterance, the first audio data corresponding to a first identifier;
perform first speech processing on the first audio data to determine results data;
receive second audio data corresponding to a second utterance, the second audio data corresponding to the first identifier;
perform second speech processing on the second audio data to determine that the second utterance requests deletion of data corresponding to the first utterance;
based at least in part on the first identifier, identify stored data corresponding to the first utterance; and
delete the stored data.

### Claim 32 (depends on 31)

The system of claim 31, wherein the second utterance directly follows the first utterance with respect to the first identifier and wherein the second speech processing determines the second utterance refers to a previous utterance.

### Claim 33 (depends on 31)

The system of claim 31, wherein the stored data comprises at least a portion of the first audio data.

### Claim 34 (depends on 31)

The system of claim 31, wherein the stored data comprises a textual representation of the first utterance.

### Claim 35 (depends on 31)

The system of claim 31, wherein the stored data comprises first data linking the first utterance to usage history data.

### Claim 36 (depends on 31)

The system of claim 31, wherein the stored data comprises the results data.

### Claim 37 (depends on 31)

The system of claim 31, wherein the first identifier corresponds to a user identifier.

### Claim 38 (depends on 31)

The system of claim 31, wherein the first identifier corresponds to a session identifier.

### Claim 39 (depends on 31)

The system of claim 31, wherein the at least one memory further comprises instructions that, when executed by the at least one processor, further cause the system to:
causing output of the results data,
wherein the second audio data is received after causing output of the results data.

### Claim 40 (depends on 31)

The system of claim 31, wherein the at least one memory further comprises instructions that, when executed by the at least one processor, further cause the system to, prior to deleting the stored data:
determine the second utterance was spoken by a same user as the first utterance.