Audio Matching System Using Client-Side Landmark Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current audio matching systems face issues with accuracy, waiting time, and computation load, particularly in noisy environments and when searching for matching audio segments.
Innovation Solution
A computer system with a client terminal and server, utilizing a classifier to estimate identification difficulty, a landmark extraction module to determine and extract landmarks, and an encoder to generate hash keys for integration and table lookup, improving matching accuracy and reducing waiting time and computation load.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional audio matching methods are used, then the system is simple to implement, but the matching accuracy is insufficient and computation load is high
Solution Approach 1:
The patent segments the audio signal into multiple frames and extracts landmarks from each frame separately. The audio signal is divided into overlapping frames, and landmarks are extracted from each frame independently, then integrated to form the final audio fingerprint. This segmentation approach improves matching accuracy by capturing local characteristics while reducing computation load through parallel processing of smaller units.
Solution Approach 2:
The patent performs preliminary landmark extraction and hash key generation on the client terminal before transmitting to the server. The client terminal pre-processes the audio signal by extracting landmarks, generating hash keys, and integrating them into an audio fingerprint. This preliminary action reduces the computation load on the server and allows for sequential data upload, decreasing waiting time.
2Measurement precision
If more computation is performed to improve accuracy, then matching precision increases, but waiting time increases
Solution Approach 1:
The patent performs preliminary landmark extraction, hash key generation, and fingerprint integration on the client terminal before transmission. This pre-processing reduces the waiting time by completing computationally intensive tasks locally before upload, while still achieving high accuracy through the sophisticated landmark extraction and integration algorithms.
Solution Approach 2:
The patent uses hash keys as simplified copies of the original audio landmarks. Instead of transmitting or comparing raw audio data or complex landmark structures, the system uses compact hash key representations that preserve the essential matching information. This copying approach maintains matching accuracy while significantly reducing computation time and data transmission requirements.
3Measurement precision
If comprehensive audio analysis is performed, then matching accuracy improves, but computation load on server increases
Solution Approach 1:
The patent shifts the computation load from the server to the client terminal by performing preliminary landmark extraction, hash key generation, and fingerprint integration before transmission. The server only needs to perform simple table lookups and comparisons, dramatically reducing server computation load while maintaining comprehensive audio analysis through the client-side processing.
Solution Approach 2:
The patent extracts only the essential matching information (landmarks and their hash keys) from the complete audio signal, discarding redundant data. The landmark extraction module identifies key features from each audio frame and converts them to hash keys, extracting only the necessary information for matching while eliminating unnecessary computation and data transmission.
Data Source
AI summary
A computer system is configured for performing audio matching. The computer system includes a client terminal and a server. The client terminal includes a classifier, a first landmark extraction module, a first encoder module and an integration module. The classifier is configured for estimating an identification difficulty. The first landmark extraction module is configured for determining the number of times for landmark extraction, and for extracting first landmarks. The first encoder module is configured for encoding the first landmarks. The server includes a second landmark extraction module, a second encoder module, a hash table, a table-lookup module, an analysis module and a decision module. The second landmark extraction module is configured for extracting second landmarks. The second encoder module is configured for encoding the second landmarks. The table-lookup module is configured for performing a table lookup. The decision module is configured for determining an audio matching result.