AUDIO CLASSIFICATION METHOD AND APPARATUS, INTELLIGENT DEVICE, AND STORAGE MEDIUM

摘要

An audio classification method and apparatus, an intelligent device, and a storage medium. The method comprises: obtaining an audio file to be classified (S11); processing an audio signal of the audio file to be classified, so as to generate an input vector indicating a first audio feature (S12), the first audio feature being an audio feature corresponding the audio file to be classified; inputting the input vector into a preset audio classification neural network model obtained by means of training (S13); and analyzing the input vector by means of the neural network model, so as to generate the classification result of the audio file to be classified (S14). By means of the method, limitations of the prior art can be broken through, and the accuracy and the efficiency for classifying audio files to be classified can be improved.