whisper_4_with_init_sun_char_0010

This model is a fine-tuned version of openai/whisper-tiny on an unknown dataset. It achieves the following results on the evaluation set:

• Train Loss: 2.8976
• Train Accuracy: 0.0339
• Train Wermet: 0.0729
• Validation Loss: 2.5997
• Validation Accuracy: 0.0245
• Validation Wermet: 0.0944
• Epoch: 9

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:

• optimizer: {'name': 'AdamWeightDecay', 'learning_rate': 1e-05, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-07, 'amsgrad': False, 'weight_decay_rate': 0.01}
• training_precision: float32

Training results

Train Loss	Train Accuracy	Train Wermet	Validation Loss	Validation Accuracy	Validation Wermet	Epoch
3.2071	0.0313	0.1237	2.8546	0.0225	0.1109	0
3.0365	0.0325	0.0375	2.8115	0.0228	0.1215	1
3.0162	0.0326	0.0484	2.7884	0.0231	0.1318	2
3.0042	0.0327	0.0555	2.7853	0.0233	0.1393	3
2.9934	0.0328	0.0614	2.7657	0.0232	0.1273	4
2.9858	0.0329	0.0654	2.7542	0.0234	0.1073	5
2.9735	0.0330	0.0673	2.7367	0.0234	0.1414	6
2.9574	0.0332	0.0704	2.6961	0.0240	0.1429	7
2.9320	0.0335	0.0723	2.6652	0.0239	0.0990	8
2.8976	0.0339	0.0729	2.5997	0.0245	0.0944	9

Framework versions

• Transformers 4.34.0.dev0
• TensorFlow 2.13.0
• Tokenizers 0.13.3