Train a sound classification model using CNN and Pytorch

There is a popular dataset on Violence Detection on Kaggle. I need someone who can train a sound classification model with that dataset to be able to predict if there is any "Violent" or "Non-violent" behaviour by hearing the sound. After importing the dataset and converting them to sound file, apply a filter and then use mel spectogram. Use Pycharm IDE and Pytorch and CNN to do the task. Use Python programming languae

Budget is AUD 30. Message me for more details.