训练声纹识别模型时,发现需要引入torchvision,找到一篇博客介绍较为详细,转载一下:https://blog.csdn.net/frighting_ing/article/details/121863387