Subtitle-Aligned Video Splitting for Smooth Dialogue Transitions

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Long English original videos are difficult to learn from due to their length and the scarcity of short video resources, and existing methods of splitting them into short videos result in abrupt dialog starts or ends, leading to poor user experience.

Innovation Solution

A method that determines video start and end points of scenes based on subtitle segment time points to smoothly transition dialogues, ensuring smooth starts and ends of short videos.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If automatic video splitting is performed using subtitle files, then productivity is improved, but manufacturing precision deteriorates due to abrupt dialog starts or ends

Engineering Contradiction:
Improvevideo splitting efficiencyVSAvoiddialog transition smoothness
Core Design Contradiction:
ProductivityVSManufacturing precision

Solution Approach 1:

The method performs preliminary analysis of subtitle segments to identify dialog boundaries before final video splitting. By pre-processing the subtitle file to extract dialog start and end time points, the system prepares optimization data in advance that enables smooth dialog transitions during the actual video splitting operation, resolving the contradiction between automated efficiency and transition quality

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The method dynamically adjusts video splitting parameters by incorporating subtitle-based time points. Instead of using fixed or purely automated splitting intervals, the system modifies the splitting parameters based on dialog boundary detection from subtitle segments, allowing the split points to align with natural dialog transitions rather than arbitrary time markers

Inventive Principle:
Principle #35Parameter changes

2Manufacturing precision

If manual video splitting is performed to achieve smooth dialog transitions, then manufacturing precision is improved, but productivity deteriorates due to high labor costs

Engineering Contradiction:
Improvedialog transition smoothnessVSAvoidvideo splitting efficiency
Core Design Contradiction:
Manufacturing precisionVSProductivity

Solution Approach 1:

The system enables self-service automated splitting by having the video processing system automatically analyze and extract dialog boundary information from subtitle files. The method allows the system to autonomously determine optimal split points based on subtitle content without requiring manual intervention, thereby achieving both high precision dialog transitions and maintained productivity

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The subtitle file serves as an intermediary that bridges automated processing and quality output. By using subtitle segments as a mediator to identify dialog boundaries, the system translates unstructured video content into structured splitting points, enabling automated systems to achieve results previously only attainable through manual analysis

Inventive Principle:
Principle #24Intermediary (Mediator)

3Device complexity

If video splitting is performed without considering subtitle timing, then device complexity is reduced, but loss of information increases due to abrupt dialog cuts

Engineering Contradiction:
Improveprocessing algorithm simplicityVSAvoiddialog continuity
Core Design Contradiction:
Device complexityVSLoss of information

Solution Approach 1:

The method performs preliminary extraction of dialog boundary information from subtitle files before video splitting. By pre-identifying dialog start and end points from subtitle segments, the system prepares continuity-preserving markers in advance, ensuring that video splitting occurs at appropriate boundaries without requiring complex real-time analysis during the splitting process

Inventive Principle:
Principle #10Preliminary action

4Ease of operation

If video splitting is performed without considering subtitle timing, then ease of operation is improved, but reliability deteriorates due to poor user experience from abrupt transitions

Engineering Contradiction:
Improvesplitting operation simplicityVSAvoiduser experience quality
Core Design Contradiction:
Ease of operationVSReliability

Solution Approach 1:

Subtitle files serve as an intermediary that guides the video splitting operation. By using subtitle segment time points as reference markers, the system maintains simple automated operation while ensuring reliable dialog transitions. The subtitle intermediary translates operational simplicity into reliable output quality by providing structured timing information

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentEP4351152B1Video splitting method and electronic device
Publication Date: 2025.08.06 PETAL CLOUD TECH CO LTD
  • EP4351152B1 patent drawingFigure 1
  • EP4351152B1 patent drawingFigure 2
  • EP4351152B1 patent drawingFigure 3

AI summary

This application provides a video splitting method and an electronic device. When the method is performed by a server, the server processes a long video into a plurality of short videos, and then a terminal obtains the short video from the server and plays the short video; or when the method is performed by a terminal, the terminal obtains a long video from a server, and then the long video is processed into a plurality of short videos, and the plurality of short videos are played. A method for processing a long video into a plurality of short videos by the server or the terminal is: first determining a video start time point and a video end time point of each scene in the long video, then determining a subtitle start time point and a subtitle end time point of each subtitle segment in a subtitle file corresponding to the long video, then adjusting the video start time point and the video end time point of the scene based on the subtitle start time point and the subtitle end time point of the subtitle segment corresponding to the scene, and finally splitting the long video based on the adjusted video start time point and the adjusted video end time point of each scene, to generate a plurality of short videos.