SOUND CLASSIFICATION METHOD, DEVICE AND MEDIUM BASED ON SEMI-NONNEGATIVE MATERIX FACTORIZATION WITH CONSTRAINT

摘要

Disclosed are a sound classification method, device and medium based on semi-nonnegative matrix factorization with constraint. The sound classification method comprises the following steps: representing a training sound data sample and a testing sound data sample as a semi-nonnegative matrix (S1); constructing a category constraint matrix and a sparse constraint matrix according to the semi-nonnegative matrix (S2); performing semi-nonnegative matrix factorization with constraint on the semi-nonnegative matrix under category constraint and sparse constraint to obtain a corresponding coefficient matrix; training a classification model to obtain a classifier by using low-dimensional representation in the coefficient matrix corresponding to the training sound data sample and category information of the training sound data sample as training data (S3); inputting the low-dimensional representation in the coefficient matrix corresponding to the testing sound data sample into the classifier, and outputting a classification result of the testing sound data sample (S4). The method makes effective use of the category information of the training sound data sample and enables the low-dimensional representation after dimension reduction to have sparsity, thereby obtaining the low-dimensional representation of samples with better discrimination and improving the accuracy of the sound data classification method.