Pattern Based Audio Searching System
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing audio indexing and retrieval methods require manual labeling, are inaccurate due to lack of context consideration, and are prone to errors from sequential retrieval strategies, limiting their applicability and accuracy.
Innovation Solution
A pattern-based audio searching method and system that automatically labels source audio data using iterative segmentation and clustering processes, builds context pattern-based decision trees, and trains segment labeling models to provide accurate retrieval results without manual labeling, considering audio class similarity and context patterns.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual labeling is used for audio indexing, then retrieval can be performed, but the process requires large amounts of manual work and is limited to special fields
Solution Approach 1:
The system performs automatic audio indexing and retrieval without manual labeling by using audio feature extraction, pattern recognition, and similarity comparison algorithms to autonomously process and search audio data
Solution Approach 2:
Manual labeling operations are replaced with automated computational processes including audio feature extraction, pattern matching algorithms, and similarity calculation systems that perform indexing and retrieval tasks
2Ease of operation
If audio retrieval is based only on audio labels, then the process is simple, but retrieval accuracy is poor due to lack of context consideration
Solution Approach 1:
The system transitions from one-dimensional audio label matching to multi-dimensional pattern recognition by incorporating temporal patterns, spectral characteristics, and contextual information to improve retrieval accuracy while maintaining operational simplicity
Solution Approach 2:
The retrieval system combines multiple types of audio features and patterns (temporal, spectral, contextual) into a composite analysis framework that achieves high accuracy without increasing operational complexity
3Device complexity
If single round sequential retrieval strategy is used, then the process is straightforward, but errors accumulate from previous steps affecting final results
Solution Approach 1:
The system performs preliminary audio feature extraction and pattern recognition before retrieval, establishing accurate baseline characteristics that prevent error accumulation in subsequent processing steps
Solution Approach 2:
The system implements feedback mechanisms where retrieval results are continuously refined by comparing against established audio patterns and adjusting matching criteria to eliminate accumulated errors and improve final accuracy
Data Source
AI summary
A pattern based audio searching method includes labeling a plurality of source audio data based on patterns to obtain audio label sequences of the source audio data; obtaining, with a processing device, an audio label sequence of target audio data; determining matching degree between the target audio data and the source audio data according to a predetermined matching rule based on the audio label sequence of the target audio data and the audio label sequences of the source audio data; and outputting source audio data having matching degree higher than a predetermined matching threshold as a search result.