Accompaniment Classification Using CNN Feature Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The uneven quality of accompaniments in the market due to limitations in recording equipment and personnel, leading to inefficient and costly manual classification processes, especially for old songs where original accompaniments are lost or of poor quality.
Innovation Solution
An accompaniment classification method using a convolutional neural network (CNN) model that extracts and normalizes audio features from target accompaniments, allowing for automatic categorization based on probability thresholds, thereby improving efficiency and reducing labor costs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual classification is used for accompaniments, then classification accuracy can be maintained, but labor costs increase and efficiency decreases
Solution Approach 1:
The patent replaces manual mechanical classification with an automated deep learning system. A pre-trained classification model processes audio features of accompaniments to automatically determine categories, substituting human labor with computational algorithms that maintain accuracy while dramatically improving efficiency and reducing costs.
Solution Approach 2:
The system enables accompaniments to be classified autonomously without human intervention. The classification model self-processes audio features and automatically outputs category results, making the classification service self-executing and eliminating the need for manual labor in the classification process.
2Manufacturing precision
If manual re-making of accompaniments is performed for old songs, then accompaniment quality is high, but cost is high and quantity is scarce
Solution Approach 1:
The patent uses silencing technology to create copies of accompaniments from existing audio recordings. By automatically separating and extracting accompaniment tracks from full songs, the system produces multiple copies of accompaniments at low cost, dramatically increasing quantity while maintaining acceptable quality through automated processing.
Solution Approach 2:
The system changes the processing parameters from manual re-making to automated silencing technology. This parameter change transforms the production method, enabling high-volume generation of accompaniments with consistent quality through automated audio separation and processing techniques.
3Productivity
If silencing technology is used for old songs, then cost is low and quantity is large, but accompaniment quality is poor
Solution Approach 1:
The patent applies preliminary classification to identify which accompaniments require quality enhancement. The system first categorizes accompaniments using the pre-trained model, then selectively applies quality improvement processing only to those identified as low quality, optimizing resource allocation while maintaining overall quality standards across large quantities.
Data Source
AI summary
An accompaniment classification method and apparatus is provided. The method includes the following. A first type of audio features of a target accompaniment is obtained (S301, S401). Data normalization is performed on each kind of audio features in the first type of audio features of the target accompaniment to obtain a first feature-set of the target accompaniment and the first feature-set is input into a first classification model for processing (S302, S402). A first probability value output by the first classification model for the first feature-set is obtained (S303, S403). An accompaniment category of the target accompaniment is determined to be a first category of accompaniments when the first probability value is greater than a first classification threshold (S404). The accompaniment category of the target accompaniment is determined to be other categories of accompaniments when the first probability value is less than or equal to the first classification threshold.