AST_EmoRecog_Model_v4

This model is a fine-tuned version of MIT/ast-finetuned-audioset-10-10-0.4593 on the IEMOCAP dataset. It achieves the following results on the evaluation set:

• Loss: 1.4615
• Accuracy: 0.5159
• Recall: 0.4007
• Precision: 0.4956
• F1: 0.4090

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 1e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 6

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Recall	Precision	F1
1.4443	1.0	377	1.3359	0.4695	0.3408	0.4793	0.3099
1.1556	2.0	754	1.2506	0.5266	0.3877	0.6026	0.3970
0.8988	3.0	1131	1.2633	0.5279	0.4175	0.5148	0.4208
0.6187	4.0	1508	1.3426	0.5279	0.4031	0.5425	0.4153
0.3944	5.0	1885	1.4266	0.5206	0.4021	0.5256	0.4152
0.2555	6.0	2262	1.4615	0.5159	0.4007	0.4956	0.4090

Framework versions

• Transformers 4.47.1
• Pytorch 2.5.1+cu121
• Datasets 3.2.0
• Tokenizers 0.21.0