Live Stream Translation Timing for Accurate Synchronized Subtitles

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Inaccurate live streaming translation results due to delayed and incomplete translation of live streams, leading to poor user experience and synchronization issues in real-time live streaming scenarios.

Innovation Solution

A live streaming translation method that acquires a candidate live stream with a target timestamp at the end of a previous stable translation result, performs translation processing, and re-encodes the stream to ensure the translation result is accurate and synchronized with the live stream, within a preset duration threshold.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Speed

If translation processing is performed on short-duration live streams in real-time, then real-time performance is improved, but translation accuracy deteriorates due to incomplete content

Engineering Contradiction:

Improvereal-time performanceVSAvoidtranslation accuracy

Core Design Contradiction:

SpeedVSMeasurement precision

Solution Approach 1:

The system performs preliminary translation processing on captured live streams with longer durations before pushing to viewers. By preparing translation results in advance for extended content segments, the system ensures complete semantic units are translated accurately while still delivering results within acceptable timeframes to maintain real-time performance.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The live stream processing is divided into captured live streams and to-be-pushed target live streams with distinct time boundaries. Translation is performed on segmented portions (captured streams) that can be processed more thoroughly, then selectively pushed as complete translation units, resolving the conflict between processing time and completeness.

Inventive Principle:

Principle #1Segmentation

2Measurement precision

If translation is delayed to ensure accuracy, then translation quality improves, but synchronization with live stream deteriorates

Engineering Contradiction:

Improvetranslation qualityVSAvoidsynchronization timing

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The system dynamically adjusts the timing and duration of captured live streams based on content completeness and translation processing status. By flexibly determining when a captured stream segment is ready (neither too early nor too late), the system maintains both translation quality and synchronization with the live stream playback timeline.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system uses timestamp feedback to track the relationship between captured live streams and to-be-pushed target live streams. By monitoring end timestamps and duration thresholds, the system ensures translation results are pushed at the correct moment to maintain synchronization while allowing sufficient processing time for quality.

Inventive Principle:

Principle #23Feedback

3Duration of action of moving object

If live stream duration is extended for complete translation, then translation completeness improves, but real-time delivery capability deteriorates

Engineering Contradiction:

Improvelive stream durationVSAvoidreal-time delivery capability

Core Design Contradiction:

Duration of action of moving objectVSProductivity

Solution Approach 1:

Translation processing is performed in advance on captured live streams that are longer than the to-be-pushed segments. This preliminary processing of extended content allows complete semantic units to be translated before delivery, ensuring translation completeness while the pre-prepared nature maintains efficient real-time delivery when pushed.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system processes more content (captured live stream with longer duration) than is immediately needed for the to-be-pushed segment. This excessive processing ensures that complete semantic units and context are available for accurate translation, while the pushed portion is selectively delivered in real-time.

Inventive Principle:

Principle #16Partial or excessive action

Data Source

PatentUS20250390688A1Live streaming translation method and apparatus, storage medium, and computer device

Publication Date: 2025.12.25 TENCENT TECHNOLOGY (SHENZHEN) CO LTD

AI summary

This application discloses a live streaming translation method performed by a computer device, including: acquiring a candidate live stream from captured live streams; performing translation processing on speech recognition content corresponding to the candidate live stream, to obtain a translation result corresponding to the candidate live stream; determining a to-be-pushed target translation result based on the translation result corresponding to the candidate live stream and a target end timestamp of a to-be-pushed target live stream; re-encoding the to-be-pushed target live stream based on the target translation result, to obtain a re-encoded live stream; and pushing the re-encoded live stream, to be displayed the target translation result at a viewer end. In this application, a duration threshold is set, so that the target translation result can be acquired and pushed within a time period corresponding to the duration threshold, to improve accuracy of a live streaming translation result.