AI-Music-Detection

This model was trained from mit/ast-finetuned-audioset-10-10-0.4593 on the SleepyJesse/aimusiclarge dataset. Please see the code in the Jupyter Notebook in files. The model was trained with `maxlength = 6000`, which is 60 seconds. This model is used to classify a given music piece is AI-generated or human-composed. The SleepyJesse/aimusiclarge dataset was used, with 80% train/test split, and `0.8` probability for audio data augmentation. See `aimusicdetectionnewlarge60.ipynb` and training metrics. The following hyperparameters were used during training: - learningrate: 5e-05 - trainbatchsize: 2 - evalbatchsize: 16 - seed: 42 - optimizer: Use OptimizerNames.ADAMWTORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizerargs=No additional optimizer arguments - lrschedulertype: linear - numepochs: 10 - Transformers 4.46.3 - Pytorch 2.5.1 - Datasets 3.1.0 - Tokenizers 0.20.3

This model was trained from mit/ast-finetuned-audioset-10-10-0.4593 on the SleepyJesse/aimusiclarge dataset. The model was trained with `maxlength = 1024`, which is 10.24 seconds. This model is used to classify a given music piece is AI-generated or human-composed. The SleepyJesse/aimusiclarge dataset was used, with 80% train/test split, and `0.8` probability for audio data augmentation. See the file `aimusicdetectionnewlarge.ipynb` and training metrics. The following hyperparameters were used during training: - learningrate: 5e-05 - trainbatchsize: 2 - evalbatchsize: 16 - seed: 42 - optimizer: Use OptimizerNames.ADAMWTORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizerargs=No additional optimizer arguments - lrschedulertype: linear - numepochs: 10 - Transformers 4.46.3 - Pytorch 2.5.1 - Datasets 3.1.0 - Tokenizers 0.20.3