Granted Patent B2

US 12,229,526 · App. 17/900,704 · Granted Feb 18, 2025

Inventors: Prakash Ghatage (Bangalore, IN); Naveen Kumar Thangaraj (Salem, IN); Kaustubh Kurhekar (Pune, IN); Sreevidya Prasad (Bangalore, IN); Sriram Sankaranarayanan (Bangalore, IN)

Assignee: ACCENTURE GLOBAL SOLUTIONS LIMITED

G06F40/58G06F16/683G06F40/263G10L15/26

Abstract

Claims (89)

1. A smart translation system, comprising:

at least one application processor;

a non-transitory processor-readable medium storing machine-readable instructions that cause the application processor to:

access input content via an application,

wherein the input content is to be translated;

extract translation metadata including a source language of the input content and a target language into which the input content is to be translated;

determine if translated content corresponding to the input content can be retrieved from a user cache;

obtain the translated content from the user cache if the translated content of the input content is in the user cache,

else if the translated content of the input content cannot be retrieved from the user cache,

determine a type of the application providing the input content, and

obtain the translated content of the input content from a translation engine automatically selected from a plurality of automatic translation engines, wherein the translated content is generated based on the type of the application;

enable a display of the translated content of the input content via the application used to display the input content wherein,

the translated content maintains visible attributes of the input content when displayed through the application.

2. The smart translation system of claim 1 , wherein to determine if the translated content of the input content is present in the user cache, the processor is to:

store prior input content submitted by a user for translation and corresponding translated content along with related translation metadata in the user cache;

calculate similarity of the input content with the prior input content; and

provide the translated content from the user cache if the similarity between the prior input content and the input content exceeds a predetermined content similarity threshold.

3. The smart translation system of claim 1 , wherein the application includes at least one of an email program, a web browser, and a software application.

4. The smart translation system of claim 3 , wherein the application includes the email program and the processor is to further:

access the input content including at least one of one or more emails and corresponding one or more attachments;

identify headers and footers separating the one or more emails;

receive a user selection of the input content to translate; and

obtain the translated content of the user selection.

5. The smart translation system of claim 4 , wherein the user selection includes the corresponding one or more attachments, and to obtain the translated content of the user selection, the processor is to further:

extract the corresponding one or more attachments from the one or more emails;

determine an associated file format of each of the corresponding one or more attachments;

identify digitized textual content of the corresponding one or more attachments; and

generate a translated file of each of the one or more attachments in the associated file format.

6. The smart translation system of claim 5 , wherein the user selection additionally includes the one of one or more emails, and to obtain the translated content of the user selection, the processor is to further:

obtain a translation of digitized textual content of the one or more emails based on the translation metadata,

wherein the translation of the digitized textual content excludes the headers and footers separating the emails; and

generate the translated content including one or more translated emails corresponding to each of the one or more emails, wherein the one or more translated emails further have attached, the one or more translated files of each of the one or more attachments.

7. The smart translation system of claim 3 , wherein the application includes the web browser, and to obtain the translated content of the input content the processor is to further:

receive the input content including user selection of at least a portion of content displayed by the web browser;

obtain the translated content of the input content based on the translation metadata; and

enable a display of the translated content of the input content in the web browser.

8. The smart translation system of claim 7 , wherein to obtain the translated content of the input content the processor is to further:

determine that the portion of the content displayed by the web browser includes an image of text;

extract the text from the image; and

obtain a translation of the text extracted from the image.

9. The smart translation system of claim 3 , wherein the application includes the software application and the processor is to further:

receive a user selection of at least a portion of the user interface of the software application; and

superimpose a context menu of default options automatically selected for the translation metadata on the user selection.

10. The smart translation system of claim 9 , wherein to obtain the translated content of the input content the processor is to further:

extract values of original fields including labels displayed on the portion of the user interface of the software application; and

obtain translated contents corresponding to each of the values.

11. The smart translation system of claim 1 , wherein the default options for the translation metadata are user-editable.

12. The smart translation system of claim 11 , wherein to enable the display of the translated content of the input content via an application, the processor is to:

generate labels including the translated contents for each of the labels with a background of the generated labels being set to match background of the original fields displayed on the portion of the user interface; and

superimpose the generated labels on the original fields.

13. The smart translation system of claim 1 , wherein the translation metadata is extracted manually or automatically.

14. The smart translation system of claim 13 , wherein to extract the translation metadata manually, the processor is to:

receive a user request for the translation metadata; and

provide a context menu that enables the user to input at least part of the translation metadata.

15. A method of providing an Artificial Intelligence (AI) based translation, comprising:

receiving input content to be translated;

extracting translation metadata including a source language of the input content and a target language into which the input content is to be translated;

determining if translated content of the input content is present in a user cache;

obtaining the translated content from the user cache if the translated content of the input content is in the user cache,

else if the translated content of the input content is not stored in the user cache,

extracting a context associated with the input content,

automatically selecting a translation engine from a plurality of automatic translation engines based on the context, and

obtaining the translated content of the input content from the selected translation engine;

enabling display of the translated content of the input content via an application used to display the input content,

wherein the translated content is displayed through the application with visible attributes similar to that of the input content.

16. The method of claim 15 , wherein receiving the input content to be translated, further includes:

receiving a user selection of content displayed on a user interface (UI) of the application.

17. The method of claim 15 , wherein extracting the context of the input content, further includes:

extracting textual content of the input content;

obtaining a domain context of the input content by providing the textual content to a machine learning (ML) domain classification model; and

obtaining a language context of the input content based on the target language.

18. The method of claim 15 , wherein obtaining the translated content of the input content from the selected translation engine further comprises:

parsing and tokenizing the input content;

identifying sensitive data from the input content based on one or more of n-grams and inference rules; and

masking the sensitive data from the input content before providing the sensitive data to the selected translation engine for translation.

19. A non-transitory processor-readable storage medium comprising machine-readable instructions that cause a processor to:

access input content to be translated via an application;

extract translation metadata including a source language of the input content and a target language into which the input content is to be translated;

determine if translated content of the input content is present in a user cache;

obtain the translated content from the user cache if the translated content of the input content is in the user cache,

else if the translated content of the input content is not stored in the user cache,

determine a type of the application providing the input content, and

obtain the translated content of the input content from a translation engine automatically selected from a plurality of automatic translation engines, wherein the translated content is generated based on the type of the application;

and

enable display of the translated content of the input content via the application used to display the input content wherein,

the translated content maintains at least one visible attribute of the input content when displayed through the application.

20. The non-transitory processor-readable storage medium of claim 19 , including further instructions that cause the processor to:

highlight key sections of the translated content in the display; and

provide a summary of the translated content.

Assignments (1)

Priority Claims (1)

Continuity (1)

References Cited (19)