Real-Time Call Translation Interface Synchronization

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current language translation during calls between users speaking different languages requires interrupting the call, involves delays, and often results in overlapping machine speech and user speech due to asynchronous translation processes.

Innovation Solution

An electronic device method that establishes a call, switches to a translation interface to display and translate speech in real-time, allowing users to change language settings during the call without interruption, and synchronizes machine speech with text translation to prevent overlap.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If language translation is performed during a call, then communication between users speaking different languages is enabled, but the call must be interrupted when language type needs to be switched

Engineering Contradiction:
Improvelanguage switching capabilityVSAvoidcall interruption time
Core Design Contradiction:
Adaptability or versatilityVSLoss of time

Solution Approach 1:

The system dynamically switches between call interface and translation interface based on user needs, allowing language type changes during the call without terminating the connection. The call connection is maintained in the background while the translation interface is activated for language switching, enabling flexible adaptation without interruption.

Inventive Principle:
Principle #15Dynamics

2Speed

If translation is performed in real-time during a call, then communication delay is reduced, but machine speech may overlap with user speech

Engineering Contradiction:
Improvetranslation speedVSAvoidspeech overlap
Core Design Contradiction:
SpeedVSObject-generated harmful factors

Solution Approach 1:

The system uses feedback mechanisms to monitor the translation process and detect when machine speech is about to overlap with user speech. Based on this feedback, the system adjusts the timing and delivery of translated speech, ensuring smooth communication without interference between user speech and machine speech.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The translation process is divided into periodic cycles where the system alternates between listening to user speech, processing translation, and delivering machine speech. This periodic action ensures that machine speech is delivered only during appropriate intervals when the user is not speaking, preventing overlap while maintaining real-time translation capability.

Inventive Principle:
Principle #19Periodic action

3Ease of operation

If translation interface is displayed during a call, then translation functionality is accessible, but the call interface is not visible

Engineering Contradiction:
Improvetranslation accessibilityVSAvoidcall interface visibility
Core Design Contradiction:
Ease of operationVSLoss of information

Solution Approach 1:

The system segments the user interface into separate call interface and translation interface, allowing each to be displayed and operated independently. The call interface remains accessible for call control functions while the translation interface provides translation functionality, enabling users to access both interfaces as needed without compromising the visibility or functionality of either.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentUS11843716B2Translation method and electronic device
Publication Date: 2023.12.12 HUAWEI TECH CO LTD
  • US11843716B2 patent drawing
  • US11843716B2 patent drawing
  • US11843716B2 patent drawing

AI summary

A translation method includes: a first electronic device establishes a call connection to a second electronic device and then displays a call interface; after receiving a first operation of a first user, the first electronic device switches from displaying the call to displaying a translation interface; when receiving a first speech of the first user in a first language, the translation interface sequentially displays at least a first text and a second text, where the first text is obtained by recognizing the first speech, and the second text is obtained by translating the first speech into a target language; and the first electronic device sends a machine speech in the target language to the second electronic device.