# US 2020/0402502 A1 · Communications utilizing multiple virtual assistant services

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

- Number: US 2020/0402502 A1
- Document ID: US20200402502A1
- Status: Published application
- Kind: A1
- Application number: 16/097,727
- Filed: 2016-06-21
- Published: 2020-12-24
- Earliest claimed: 2016-06-21
- Assignees: Hewlett-Packard Development Company, L.P.
- Applicants: Hewlett-Packard Development Company, L.P.
- Inventors: David H. Hanes; John Michael Main; Jon R. Dory
- Web page: https://patentgrep.com/patent/US20200402502A1

## 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 16/3329: Natural language query formulation
- G10L 15/32: Multiple recognisers used in sequence or in parallel; Score combination systems therefor, e.g. voting systems
- 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

In one example of the disclosure, microphone data indicative of a user spoken phrase is captured utilizing the microphone at a communication apparatus. At least a portion of the microphone data is sent to a set of computing devices. A response phrase determined at a virtual assistant service is received from each of the computing devices. A preferred response phrase is identified among the set of received response phrases according to a preference rule. The preferred response phrase is caused to be output via a speaker at the communication apparatus.

## Claims

### Claim 1

A communication apparatus, comprising:
a speaker;
a microphone;
a phrase capture engine, to capture, utilizing the microphone, microphone data indicative of a user spoken phrase;
a microphone data sending engine, to send at least a portion of the microphone data to a set of computing devices, wherein each of the computing devices is to communicate with a server system hosting a distinct virtual assistant service;
a response receipt engine, to receive from each of the computing devices a response phrase determined at a virtual assistant service;
a response comparison engine, to identify a preferred response phrase among the set of received response phrases according to a preference rule; and
a response output engine to cause output, via the speaker, of the preferred response phrase.

### Claim 2 (depends on 1)

The communication apparatus of claim 1, wherein the response comparison engine is to identify the preferred response phrase by assigning scores to the received response phrases according to the preference rule.

### Claim 3 (depends on 1)

The communication apparatus of claim 1, wherein according to the preference rule a received response phrase determined to be successful in providing information relevant to the user spoken phrase is assigned a higher preference than a received response phrase that has been determined as failing to provide information relevant to the user spoken phrase.

### Claim 4 (depends on 3)

The communication apparatus of claim 3, wherein success or failure of a received response phrase to provide information relevant to the user spoken phrase is determined based upon a salient response algorithm.

### Claim 5 (depends on 1)

The communication apparatus of claim 1, wherein according to the preference rule a received response phrase indicative of an error message is to be assigned a lower preference than a preference that would be assigned to a received response phrase not indicative of an error message.

### Claim 6 (depends on 1)

The communication apparatus of claim 1, wherein the preference rule includes as a factor past predictive success of virtual assistant services with respect to a subject category.

### Claim 7 (depends on 1)

The communication apparatus of claim 1, further comprising a user preference engine to capture preference data indicative of a user preference for a first virtual assistant service, and wherein the response comparison engine is to identify the preferred response phrase in consideration of the user preference.

### Claim 8 (depends on 7)

The communication apparatus of claim 7, wherein the preference data is indicative of a user preference for a first virtual assistant service for a subject category, and wherein the response comparison engine is to identify the preferred response phrase in consideration of the user preference.

### Claim 9

A memory resource storing instructions that when executed cause a processing resource to effect communication between a communication apparatus and a computing device, the instructions comprising:
a phrase capture module that when executed causes the processing resource to capture at a communication apparatus including a speaker and a microphone, microphone data indicative of a user spoken phrase;
a microphone data sending module that when executed causes the processing resource to send at least a portion of the microphone data to a set of server systems, wherein each server system hosts a distinct virtual assistant service;
a response receipt module that when executed causes the processing resource to receive from each of the server systems a response phrase determined at a virtual assistant service;
a response comparison module that when executed causes the processing resource to identify a preferred response phrase among the set of received response phrases according to a preference rule; and
a response output module that when executed causes the processing resource to cause output, via the speaker, of the preferred response phrase.

### Claim 10 (depends on 9)

The computer readable medium of claim 9, wherein the response comparison module when executed causes the processing resource to identify the preferred response phrase by assigning scores to the received response phrases according to the preference rule.

### Claim 11 (depends on 9)

The computer readable medium of claim 9, wherein according to the preference rule a received response phrase that has a higher saliency rating according to a salient response algorithm is to be assigned a higher score than a received response phrase that has a lower saliency rating.

### Claim 12 (depends on 9)

The computer readable medium of claim 9, wherein according to the preference rule a received response phrase that is indicative of an error, or that has an associated error message or error code, is to be assigned a lower score than a score that would be assigned to a received response phrase that is not indicative of an error message.

### Claim 13 (depends on 9)

The computer readable medium of claim 9, further comprising a user preference module that when executed causes the processing resource to capture preference data indicative of a user preference for a first virtual assistant service, and wherein the response comparison module when executed causes the processing resource to identify the preferred response phrase in consideration of the user preference.

### Claim 14 (depends on 13)

The computer readable medium of claim 13, wherein the preference data is indicative of a user preference for a first virtual assistant service for a subject category, and wherein the response comparison module when executed causes the processing resource to identify the preferred response phrase in consideration of the user preference.

### Claim 15

A communication method, comprising:
capturing at a communication apparatus microphone data, the microphone data including a user spoken phrase;
sending at least a portion of the microphone data to a set of computing devices, wherein each of the computing devices is to communicate with a server system hosting a distinct virtual assistant service;
receiving from each of the computing devices a response phrase determined at a virtual assistant service;
identifying a preferred response phrase among the set of received response phrases by assigning a preference to each of the received response phrases according to a preference rule, wherein the preference rule includes as a factor at least one from the set of
success or failure of a received response phrase determined based upon a salient response algorithm,
presence of an error message in a received response;
past predictive success of a virtual assistant service with respect to a subject category, and
a user preference for a virtual assistant service for a subject category; and
causing output, via the speaker, of the preferred response phrase.
