IP Library Granted Patent US 12,591,421
Granted Patent B2
US 12,591,421 · App. 18/530,055 · Granted Mar 31, 2026

Systems and methods for enabling software code for different languages

Inventors: Michail Vasiltschenko (Schriesheim, DE); Thomas Wienold (Offenbach, DE); Christian Lieske (Malsch, DE); Maite Friedrich Dupont (Porto Alegre, BR)
Assignee: SAP SE
G06F8/51G06F11/3604G06F40/279
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,591,421
App. No.
18/530,055
Granted
Mar 31, 2026
Kind
B2
Abstract

Embodiments of the present disclosure include techniques for enabling software code for different languages. In one embodiment, a user creates a prompt. The prompt may be to generate new code or modify existing code. The prompt may be automatically modified to include one or more instructions to extract code elements from source code and the code elements may be translated. In some embodiments, translation services are incorporated into and IDE or build environment as a plugin.

Claims (36)

1 . A method of generating code in a plurality of languages comprising:

entering a prompt specifying source code and an instruction to extract translatable code elements from the source code into a generative large language model;

receiving, from the generative large language model, a first plurality of code elements extracted from the source code comprising first hard coded text strings output to a user;

sending the first plurality of code elements to a multi-language software system for translating the first hard coded text strings to second hard coded text strings in a different natural language than the first hard coded text strings, wherein the generative large language model is trained to identify the first hard coded text strings and replace the first hard coded text strings with links to one or more external files comprising the second hard coded text strings;

receiving, from the multi-language software system, a second plurality of code elements comprising the second hard coded text strings; and

replacing the first plurality of code elements in the source code with the second plurality of code elements.

2 . The method of claim 1 , wherein the prompt is generated by a user.

3 . The method of claim 1 , wherein the prompt is a second prompt and the source code is source code to be generated, the method further comprising receiving a first prompt from a user to generate the source code and automatically modifying the second prompt to include the instruction to extract translatable code elements from the source code.

4 . The method of claim 1 , wherein the multi-language software system is a language translation software system that translates the first plurality of code elements in a first natural language to the second plurality of code elements in the different natural language.

5 . The method of claim 1 , further comprising:

reviewing, by a user, the second plurality of code elements; and

entering the second plurality of code elements into a language verification software system to verify the second plurality of code elements.

6 . The method of claim 1 , wherein the entering, receiving the first plurality of code elements, sending, receiving the second plurality of code elements, and replacement steps are incorporated into an integrated development environment as a translation plug-in.

7 . The method of claim 1 , wherein the entering, receiving the first plurality of code elements, sending, receiving the second plurality of code elements, and replacement steps are incorporated into a build environment as a translation plug-in.

8 . The method of claim 1 , wherein the prompt specifying source code comprises an instruction to generate source code.

9 . The method of claim 1 , wherein the prompt specifying source code comprises the source code.

10 . The method of claim 1 , wherein the instruction to extract translatable code elements from the source code specifies a file to store the first plurality of code elements.

11 . The method of claim 1 , wherein the prompt specifies a data structure of the source code and text values to be extracted.

12 . A computer system comprising:

at least one processor;

at least one non-transitory computer readable medium storing computer executable instructions that, when executed by the at least one processor, cause the computer system to perform operations for generating code in a plurality of languages comprising:

entering a prompt specifying source code and an instruction to extract translatable code elements from the source code into a generative large language model;

receiving, from the generative large language model, a first plurality of code elements extracted from the source code comprising first hard coded text strings output to a user;

sending the first plurality of code elements to a multi-language software system for translating the first hard coded text strings to second hard coded text strings in a different natural language than the first hard coded text strings, wherein the generative large language model is trained to identify the first hard coded text strings and replace the first hard coded text strings with links to one or more external files comprising the second hard coded text strings;

receiving, from the multi-language software system, a second plurality of code elements comprising the second hard coded text strings; and

replacing the first plurality of code elements in the source code with the second plurality of code elements.

13 . The computer system of claim 12 , wherein the instruction to extract translatable code elements from the source code specifies a file to store the second plurality of code elements in a different language.

14 . The computer system of claim 12 , wherein the prompt specifies a data structure of the source code and text values to be extracted.

15 . A non-transitory computer-readable medium storing computer-executable instructions that, when executed by at least one processor, perform operations for generating code in a plurality of languages comprising:

entering a prompt specifying source code and an instruction to extract translatable code elements from the source code into a generative large language model;

receiving, from the generative large language model, a first plurality of code elements extracted from the source code comprising first hard coded text strings output to a user;

sending the first plurality of code elements to a multi-language software system for translating the first hard coded text strings to second hard coded text strings in a different natural language than the first hard coded text strings, wherein the generative large language model is trained to identify the first hard coded text strings and replace the first hard coded text strings with links to one or more external files comprising the second hard coded text strings;

receiving, from the multi-language software system, a second plurality of code elements comprising the second hard coded text strings; and

replacing the first plurality of code elements in the source code with the second plurality of code elements.

16 . The non-transitory computer-readable medium of claim 15 , wherein the instruction to extract translatable code elements from the source code specifies a file to store the second plurality of code elements in a different natural language.

17 . The non-transitory computer-readable medium of claim 15 , wherein the prompt specifies a data structure of the source code and text values to be extracted.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Dec 7, 2023
From: WIENOLD, THOMAS; VASILTSCHENKO, MICHAIL; LIESKE, CHRISTIAN
To: SAP SE
Reel/Frame 065799/0132 →
Continuity (1)
Related Publication 20250181336A1 · Jun 5, 2025
References Cited (6)
US 10635566B1 · Talluri · 2020 [cited by examiner]
US 11693637B1 · Singh · 2023 [cited by examiner]
US 20190379770A1 · Thantry · 2019 [cited by examiner]
US 20220207246A1 · Ha · 2022 [cited by examiner]
US 20240211224A1 · Clement · 2024 [cited by examiner]
US 20250173525A1 · Verma · 2025 [cited by examiner]