# US 2024/0202468 A1 · Automated translations for autonomous chat agents

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

- Number: US 2024/0202468 A1
- Document ID: US20240202468A1
- Status: Published application
- Kind: A1
- Application number: 18/081,811
- Filed: 2022-12-15
- Published: 2024-06-20
- Earliest claimed: 2022-12-15
- Assignees: ADP, Inc.
- Applicants: ADP, Inc.
- Inventors: Sathya Vikram Chekuri; Henry Charles Will, IV
- Web page: https://patentgrep.com/patent/US20240202468A1

## CPC

- G06F 40/58: Use of machine translation, e.g. for multi-lingual retrieval, for server-side translation for client devices or for real-time translation
- G06F 40/30: Semantic analysis
- G06F 40/35: Discourse or dialogue representation
- G06F 40/45: Example-based machine translation; Alignment
- G06N 3/044: Recurrent networks, e.g. Hopfield networks
- G06N 3/0464: Convolutional networks [CNN, ConvNet]
- G06N 5/01: Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound
- G06N 20/00: Machine learning
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS

## Abstract

Disclosed are various embodiments for automated translations for autonomous chat agents. A build service can send a translation request to a machine translation service, the translation request comprising training data in a first language and the translation request specifying a second language. The build service can then receive translated training data from the machine translation service, the translated training data having been translated from the training data into the second language. Next, the build service can create a translated workflow that comprises a translated machine learning model and a translated intent. Subsequently, the build service can add the translated training data to the translated workflow and train the translated machine learning model using the translated training data.

## Claims

### Claim 1

A system, comprising:
a computing device comprising a processor and a memory; and
machine-readable instructions stored in the memory that, when executed by the processor, cause the computing device to at least:
send a translation request to a machine translation service, the translation request comprising training data in a first language and the translation request specifying a second language;
receive translated training data from the machine translation service, the translated training data having been translated from the training data into the second language;
create a translated workflow that comprises a translated machine learning model and at least one translated intent; and
add the translated training data to the translated workflow; and
train the translated machine learning model using the translated training data.

### Claim 2 (depends on 1)

The system of claim 1, wherein the machine-readable instructions further cause the computing device to at least deploy an autonomous chat agent service in the second language, the autonomous chat agent being configured to use the translated machine learning model to process user requests in the second language.

### Claim 3 (depends on 1)

The system of claim 1, wherein the machine-readable instructions further cause the computing device to at least:
receive an override instruction for the intent, the override instruction specifying a replacement value for at least one word or phrase in the translated training data; and
include the replacement value in the translated training data.

### Claim 4 (depends on 1)

The system of claim 1, wherein the translation request further comprises one or more responses in the first language and the machine-readable instructions further cause the computing device to at least receive one or more translated responses from the machine translation service, the translated responses having been translated from the one or more responses into the second language.

### Claim 5 (depends on 1)

The system of claim 1, wherein the translation request is a first translation, the workflow is a first workflow request, and the machine-readable instructions further cause the computing device to at least:
send a second translation request to the machine translation service, the second translation request comprising updated training data in the first language and the second translation request specifying the second language;
receive updated translated training data from the machine translation service, the updated translated training data having been translated from the updated training data into the second language;
create a second translated workflow to replace the first translated workflow, the second translated workflow comprising a second translated machine learning and at least one translated intent;
add the updated translated training data to the second translated workflow; and
train the second translated machine learning model using the updated translated training data.

### Claim 6 (depends on 5)

The system of claim 5, wherein the autonomous chat agent service is a first autonomous chat agent service and the machine-readable instructions further cause the computing device to at least replace the first autonomous chat agent service with a second autonomous chat agent service in the second language, the second autonomous chat agent being configured to use the second translated machine learning model to process user requests in the second language.

### Claim 7 (depends on 1)

The system of claim 1, wherein the machine-readable instructions further cause the computing device to at least add a second translated intent to the translated machine learning model, the second translated intent being based at least in part on the second language or a geographic area associated with the translated machine learning model.

### Claim 8

A method, comprising:
sending a translation request to a machine translation service, the translation request comprising training data in a first language and the translation request specifying a second language;
receiving translated training data from the machine translation service, the translated training data having been translated from the training data into the second language;
creating a translated workflow that comprises a translated machine learning model and at least one translated intent; and
adding the translated training data to the translated workflow; and
training the translated machine learning model using the translated training data.

### Claim 9 (depends on 8)

The method of claim 8, further comprising deploying an autonomous chat agent service in the second language, the autonomous chat agent being configured to use the translated machine learning model to process user requests in the second language.

### Claim 10 (depends on 8)

The method of claim 8, further comprising
receiving an override instruction for the intent, the override instruction specifying a replacement value for at least one word or phrase in the translated training data; and
including the replacement value in the translated training data.

### Claim 11 (depends on 8)

The method of claim 8, wherein the translation request further comprises one or more responses in the first language and the method further comprises receiving one or more translated responses from the machine translation service, the translated responses having been translated from the one or more responses into the second language.

### Claim 12 (depends on 8)

The method of claim 8, wherein the translation request is a first translation, the workflow is a first workflow request, and the method further comprises:
sending a second translation request to the machine translation service, the second translation request comprising updated training data in the first language and the second translation request specifying the second language;
receiving updated translated training data from the machine translation service, the updated translated training data having been translated from the updated training data into the second language;
creating a second translated workflow to replace the first translated workflow, the second translated workflow comprising a second translated machine learning and at least one translated intent;
adding the updated translated training data to the second translated workflow; and
training the second translated machine learning model using the updated translated training data.

### Claim 13 (depends on 12)

The method of claim 12, wherein the autonomous chat agent service is a first autonomous chat agent service and the method further comprises replacing the first autonomous chat agent service with a second autonomous chat agent service in the second language, the second autonomous chat agent being configured to use the second translated machine learning model to process user requests in the second language.

### Claim 14 (depends on 8)

The method of claim 8, further comprising adding a second translated intent to the translated workflow, the second translated intent being based at least in part on the second language or a geographic area associated with the translated machine learning model.

### Claim 15

A non-transitory, computer-readable medium, comprising machine-readable instructions that, when executed by a processor of a computing device, cause the computing device to at least:
send a translation request to a machine translation service, the translation request comprising training data in a first language and the translation request specifying a second language;
receive translated training data from the machine translation service, the translated training data having been translated from the training data into the second language;
create a translated workflow that comprises a translated machine learning model and at least one translated intent; and
add the translated training data to the translated workflow; and
train the translated machine learning model using the translated training data.

### Claim 16 (depends on 15)

The non-transitory, computer-readable medium of claim 15, wherein the machine-readable instructions further cause the computing device to at least deploy an autonomous chat agent service in the second language, the autonomous chat agent being configured to use the translated machine learning model to process user requests in the second language.

### Claim 17 (depends on 15)

The non-transitory, computer-readable medium of claim 15, wherein the machine-readable instructions further cause the computing device to at least
receive an override instruction for the intent, the override instruction specifying a replacement value for at least one word or phrase in the translated training data; and
include the replacement value in the translated training data.

### Claim 18 (depends on 15)

The non-transitory, computer-readable medium of claim 15, wherein the translation request further comprises one or more responses in the first language and the machine-readable instructions further cause the computing device to at least receive one or more translated responses from the machine translation service, the translated responses having been translated from the one or more responses into the second language.

### Claim 19 (depends on 15)

The non-transitory, computer-readable medium of claim 15, wherein the translation request is a first translation, the workflow is a first workflow request, and the machine-readable instructions further cause the computing device to at least:
send a second translation request to the machine translation service, the second translation request comprising updated training data in the first language and the second translation request specifying the second language;
receive updated translated training data from the machine translation service, the updated translated training data having been translated from the updated training data into the second language;
create a second translated workflow to replace the first translated workflow, the second translated workflow comprising a second translated machine learning and at least one translated intent;
add the updated translated training data to the second translated workflow; and
train the second translated machine learning model using the updated translated training data.

### Claim 20 (depends on 19)

The non-transitory, computer-readable medium of claim 19, wherein the autonomous chat agent service is a first autonomous chat agent service and the machine-readable instructions further cause the computing device to at least replace the first autonomous chat agent service with a second autonomous chat agent service in the second language, the second autonomous chat agent being configured to use the second translated machine learning model to process user requests in the second language.
