Audio data processing method and device

The invention discloses an audio data processing method and device, wherein the method comprises the steps: obtaining each piece of audio data collected in history, and determining the human voice data in each piece of audio data; and then, determining each piece of sliding window data contained in the human voice data, performing audio feature extraction on each piece of sliding window data in each piece of human voice data, inputting the extracted audio features into a voice classification model, and determining the probability that the sliding window data belongs to normal human voice; andfinally, based on the probability that each piece of sliding window data in the human voice data belongs to normal human voice, determining the normal human voice data in the human voice data so as todetermine training samples. According to the audio data processing method and device, through the method that the audio features of the audio data are extracted and the voice classification is performed based on the extracted