IP Library Granted Patent US 12706090
Granted Patent B2
US 12706090 · App. 18/599,553 · Granted Aug 11, 2026

Method and apparatus with real-time translation

Inventors: Youngmin Kim (Hwaseong-si, KR); Hwidong Na (Seongnam-si, KR); Min-Joong Lee (Suwon-si, KR); Hodong Lee (Yongin-si, KR)
Assignee: Samsung Electronics Co., Ltd.
G10L15/183G10L15/005
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 12706090
App. No.
18/599,553
Granted
Aug 11, 2026
Kind
B2
Abstract

A real-time processor-implemented translation method and apparatus is provided. The real-time translation method includes receiving a content, determining a delay time for real-time translation based on a silence interval of the received content and an utterance interval of the received content, generating a translation result by translating a language used in the received content, and synthesizing the translation result and the received content.

Claims (19)

1 . A processor-implemented translation method comprising:

receiving, by a processor, a content;

determining, by the processor, a time-shift of the received content based on a source language of the received content and a table, wherein the table has time-shift values corresponding to a plurality of languages;

determining, by the processor, a delay time for a real-time translation based on the time-shift of the received content and a silence interval in the received content;

generating, by the processor, a translation result through a trained automated translation model by translating a phrasing used in the received content based on the delay time;

synthesizing, by the processor, the generated translation result and the received content; and

providing a user with the synthesized result of the generated translation result and the received content in real-time.

2 . The method of claim 1 , wherein

the time-shift is determined based on the source language and a target language; and

the target language corresponds to a language that a user desires to view or listen.

3 . The method of claim 1 , wherein the table comprises experimentally determined time-shift values for each of the plurality of languages.

4 . The method of claim 1 , further comprising:

storing an amount of the received content corresponding to the time-shift.

5 . The method of claim 1 , wherein the determining the delay time comprises:

extracting the silence interval from the received content; and

determining the delay time based on a time corresponding to the silence interval.

6 . The method of claim 1 , wherein the generating of the translation result comprises:

generating the translation result by directly translating a speech included in the received content into speech in a target language; or

generating the translation result by translating the speech included in the received content into a text in the target language.