Robust Keyword Extraction from Noisy Voice Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing keyword extraction technologies fail to effectively extract keywords from voice recognition results that include errors due to background noise, excessive reverberation, or speaker pauses.
Innovation Solution
A keyword extraction device that utilizes a keyword extraction model, a voice recognition unit, and a word vector representation generating unit to generate and input word vector representations in descending order of reliability, allowing robust keyword extraction even from erroneous voice recognition results.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If keyword extraction is performed on voice recognition results including errors, then keyword extraction can be performed on all input data, but extraction accuracy deteriorates due to recognition errors from background noise, reverberation, or speaker pauses
Solution Approach 1:
The patent changes the parameter of word vector representation to include reliability information from voice recognition. By modifying the input parameters to the keyword extraction model to include both word vectors and reliability scores, the system can distinguish between high-confidence and low-confidence recognition results, thereby maintaining extraction accuracy even when processing voice data with errors.
Solution Approach 2:
The patent introduces reliability information as an intermediary element between voice recognition and keyword extraction. This intermediary component allows the keyword extraction model to weigh different candidate words based on their recognition reliability, effectively mediating the impact of voice recognition errors on keyword extraction accuracy.
2Device complexity
If traditional keyword extraction models are used without reliability information, then the model structure remains simple, but the model cannot handle voice recognition errors effectively
Solution Approach 1:
The patent makes the keyword extraction model multi-functional by enabling it to process both word vector representations and reliability information from voice recognition. This universal approach allows the same model to handle various input types and robustness requirements without needing separate specialized models, thereby improving reliability while maintaining reasonable structural complexity.
Data Source
AI summary
A keyword is extracted robustly despite a voice recognition result including an error. A model storage unit 10 stores a keyword extraction model that accepts word vector representations of a plurality of words as an input and extracts and outputs a word vector representation of a word to be extracted as a keyword. A speech detection unit 11 detects a speech part from a voice signal. A voice recognition unit 12 executes voice recognition on the speech part of the voice signal and outputs a confusion network which is a voice recognition result. A word vector representation generating unit 13 generates a word vector representation including reliability of voice recognition with regard to each candidate word for each confusion set. A keyword extraction unit 14 inputs the word vector representation of the candidate word to the keyword extraction model in descending order of the reliability and obtains the word vector representation of the keyword.