Slim Self-Similarity Matrix for Dense Music Ornamentation Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing digital audio signal processing techniques fail to effectively detect and synchronize multimedia with dense ornamentation in music, such as rapid drum beats or guitar solos, due to their focus on global patterns rather than local variations, and the computational expense of high-resolution self-similarity matrices.
Innovation Solution
The use of a slim self-similarity matrix that focuses on elements near the diagonal, reducing processing workload by ignoring off-diagonal elements, combined with a lower resolution full SSM for global patterns, allows for efficient detection of dense ornamentation and synchronization with multimedia responses.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a full high-resolution self-similarity matrix is used to detect dense ornamentation, then detection precision is improved, but computational complexity and processing time increase significantly
Solution Approach 1:
The patent segments the self-similarity matrix into two distinct parts: a slim band around the diagonal (capturing local dense ornamentation) and the full matrix (capturing global patterns). This segmentation allows each part to be processed at appropriate resolutions without the computational burden of processing the entire high-resolution matrix, thus resolving the contradiction between detection precision and computational complexity.
Solution Approach 2:
The patent applies different processing qualities to different regions of the matrix: high-resolution processing is applied only to the slim band around the diagonal where dense ornamentation appears, while lower-resolution processing is applied to the rest of the matrix for global patterns. This local quality approach maintains detection precision where needed while reducing overall computational complexity.
2Measurement precision
If a full high-resolution self-similarity matrix is used to detect dense ornamentation, then detection precision is improved, but processing time increases significantly
Solution Approach 1:
The patent segments the self-similarity matrix into a slim band around the diagonal and the rest of the matrix, processing only the relevant slim band at high resolution to detect dense ornamentation. This segmentation dramatically reduces the number of computations required, thereby reducing processing time while maintaining detection precision for the target feature.
Solution Approach 2:
The patent extracts only the necessary portion (the slim band around the diagonal) from the full self-similarity matrix for high-resolution processing. By taking out only the relevant data needed for detecting dense ornamentation, the processing time is reduced significantly while preserving detection precision for the target feature.
3Productivity
If existing algorithms focus on global components of music, then processing efficiency is improved, but detection of local dense ornamentation is lost
Solution Approach 1:
The patent segments the analysis into two parallel processes: one focusing on the slim band around the diagonal to detect local dense ornamentation, and another using the full matrix at lower resolution to capture global patterns. This segmentation ensures that neither local nor global information is lost, while maintaining processing efficiency through selective high-resolution processing.
Solution Approach 2:
The patent applies local quality by directing high-resolution analysis specifically to regions where dense ornamentation is likely to appear (around the diagonal), while using lower-resolution analysis for global pattern recognition. This approach prevents loss of local information while maintaining overall processing efficiency.
Data Source
AI summary
Techniques are disclosed for automatic detection of dense ornamentation in music. Input data representing a piece of digitally encoded music in a time domain is converted into a spectrogram representing time-frequency coefficients in a frequency domain. The spectrogram includes column vectors of the time-frequency coefficients that correspond to time periods spanning different portions of the piece of music. A one-dimensional onset detection array is calculated based on a subset of the column vectors. Using the spectrogram and the onset detection array, a two-dimensional self-similarity matrix (SSM) is calculated based on pair-wise comparisons of elements in the onset detection array. As a result, an irregular pattern score representing the presence of dense ornamentation in the piece of music can be calculated based on a magnitude difference between a beat pattern in the music and each column of the slim SSM.