Abstract:Clustering algorithms have the characteristics of being simple and efficient and can complete calculations without a large number of datasets, making them suitable for application in noise reduction processing for audio module mass production testing. In order to solve the problems of the NMF algorithm easily getting stuck in local optimal solutions and difficult feature signal extraction, an improved NMF audio denoising algorithm combined with K-means initialization was designed. Firstly, the Euclidean distance formula of K-means has been improved to extract audio signal features from multiple dimensions. Combined with the initialization strategy of K-means decomposition, the initialization dictionary matrix of the NMF algorithm has been optimized to avoid getting stuck in local optimal solutions and effectively improve the robustness of the algorithm. Secondly, in the sparse coding part of the NMF algorithm, feature extraction expressions are added to solve the problem of noise residue and partial spectral signal loss in audio signals during the operation process. At the same time, the size of the coefficient matrix is limited to reduce operation time and improve the accuracy of feature extraction in high-precision audio signals. Then, comparative experiments were conducted using the NOIZEUS and NOISEX-92 datasets, as well as random noise audio signals. This algorithm improved the signal-to-noise ratio by 10–20 dB and reduced harmonic distortion by approximately −10 dB. Finally, a high-precision audio acquisition unit based on FPGA was designed, and practical applications have shown that it can effectively improve the signal-to-noise ratio of audio signals and reduce harmonic distortion.

Online algorithms for Nonnegative Matrix Factorization with the Itakura-Saito divergence

Online Nonnegative Matrix Factorization with Robust Stochastic Approximation.

Non-negative matrix factorization with α-divergence

Online Adaptive Passive-Aggressive Methods for Non-Negative Matrix Factorization and Its Applications

Algorithms for Non-Negative Matrix Factorization on Noisy Data With Negative Values

Optimization and expansion of non-negative matrix factorization

Fast Parallel Randomized Algorithm for Nonnegative Matrix Factorization with KL Divergence for Large Sparse Datasets

A High-Performance Parallel Algorithm for Nonnegative Matrix Factorization

Rethinking Non-Negative Matrix Factorization with Implicit Neural Representations

Orthogonal Nonnegative Matrix Factorization with the Kullback-Leibler divergence

Factor-Bounded Nonnegative Matrix Factorization

Randomized Algorithms for Symmetric Nonnegative Matrix Factorization

Nonnegative Matrix Factorization on Orthogonal Subspace

Applications of Online Nonnegative Matrix Factorization to Image and Time-Series Data

Convergent Projective Non-Negative Matrix Factorization with Kullback-Leibler Divergence

An Improved Nonnegative Matrix Factorization Algorithm Combined with K-Means for Audio Noise Reduction

A Gradient-Based Algorithm with Nonmonotone Line Search for Nonnegative Matrix Factorization

NeNMF: an Optimal Gradient Method for Nonnegative Matrix Factorization.

Efficient algorithms for regularized Poisson Non-negative Matrix Factorization

Non-negative Matrix-Set Factorization

Non-negative matrix factorization and deconvolution as dual simplex problem