Automatic Role Labeling for Multi-Character Audio Novels
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Manual role labeling in multi-character audio novels is time-consuming and costly, with low efficiency due to the need for manual identification of dialogue sentences and their corresponding roles.
Innovation Solution
A method that involves obtaining dialogue sentences and context information, splicing them to create a spliced text, extracting location information of role names, determining first and second candidate role names, and using these to automatically label the dialogue sentences based on a target role name.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual role labeling is used to identify dialogue sentences and their corresponding roles, then labeling accuracy can be maintained, but time consumption and labor costs increase significantly
Solution Approach 1:
The system enables automatic role labeling where the computer automatically identifies dialogue sentences and extracts role names without human intervention. The automated system processes text to determine candidate role names and selects target role names, replacing manual staff work while maintaining acceptable accuracy through algorithmic analysis of context and position information.
Solution Approach 2:
The patent replaces the mechanical manual process of reading and labeling with an automated computational system. The system uses algorithms to extract location information, generate candidate role names, and determine target role names automatically, substituting human cognitive work with machine-based text processing and pattern recognition.
2Measurement precision
If manual role labeling is performed by staff reading full text, then role names can be accurately identified, but labor costs and operational complexity increase
Solution Approach 1:
The labeling process is segmented into distinct automated stages: extracting location information of potential role names, generating first candidate role names based on position, identifying second candidate role names from the text, and selecting target role names. This segmentation allows each step to be handled by specific algorithms, reducing overall operational complexity while maintaining accuracy.
Solution Approach 2:
The system introduces intermediate processing steps between raw text and final labels, including extraction of location information and generation of candidate role names. These intermediaries structure the data in a way that simplifies the final selection process, reducing the complexity of direct manual annotation while preserving accuracy.
3Productivity
If automatic role labeling is implemented, then time consumption and labor costs are reduced, but labeling accuracy may deteriorate
Solution Approach 1:
The system generates multiple candidate role names (both first and second candidates) rather than directly selecting a single label. This excessive action of creating multiple possibilities allows the system to evaluate different options and select the most accurate target role name, improving accuracy while maintaining automated efficiency.
Solution Approach 2:
The system uses location information and context from the text to feedback into the candidate generation and selection process. By continuously refining candidate role names based on their positional context and textual evidence, the system improves labeling accuracy through iterative automated evaluation rather than single-pass labeling.
Data Source
AI summary
The disclosure provides a role labeling method. The method includes: obtaining a dialogue sentence to be labeled and context information corresponding to the dialogue sentence, and splicing the context information and the dialogue sentence to obtain a spliced text; extracting location information of a role name of the dialogue sentence in the spliced text from the spliced text; determining a first candidate role name of the dialogue sentence based on the location information; determining a second candidate role name of the dialogue sentence from role names in the spliced text; and determining a target role name of the dialogue sentence based on the first candidate role name and the second candidate role name, and performing role labeling on the dialogue sentence based on the target role name.