CngFSt10_5_sec_model

This model is a fine-tuned version of openai/whisper-medium on the Marcusxx/CngFSt10_5 dataset. It achieves the following results on the evaluation set:

• Loss: 0.0061
• Cer: 15.8432

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: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 250
• training_steps: 3000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Cer
0.0207	3.6765	1000	0.0354	36.5029
0.0007	7.3529	2000	0.0076	15.6500
0.0003	11.0294	3000	0.0061	15.8432

Framework versions

• Transformers 4.41.2
• Pytorch 2.2.2+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1