Abstract: To solve the problems of low efficiency and high experience requirement in manual detection of the abnormal sound generated by electric shaver blade rotation, an acoustic testing method combining the wavelet transform and the support vector machine (SVM) optimized by artificial fish swarm algorithm is proposed. Firstly, the acoustic signal of the electric shaver blade rotation is decomposed and reconstructed by discrete wavelet transform, and the obtained relative wavelet energy of each layer is used as the characteristic parameters of samples. Secondly, the support vector machine is optimized by the artificial fish swarm algorithm, and the optimized model is used to train samples and perform classification. The research results show that the support vector machine which is optimized by artificial fish swarm algorithm is superior to the traditional support vector machine in terms of recognition accuracy, and the recognition rate of samples reaches 95%.