whisper_charsplit_new_0008

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: 0.2282
• Train Accuracy: 0.0743
• Train Wermet: 11.7308
• Validation Loss: 0.3159
• Validation Accuracy: 0.0752
• Validation Wermet: 9.2086
• Epoch: 7

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
0.8733	0.0602	13.0686	0.6470	0.0676	11.4066	0
0.5740	0.0666	12.7778	0.5113	0.0706	11.1022	1
0.4553	0.0692	12.2404	0.4371	0.0723	10.9105	2
0.3813	0.0708	11.9157	0.3935	0.0733	9.4615	3
0.3292	0.0720	11.5732	0.3630	0.0740	9.9885	4
0.2886	0.0729	11.5171	0.3403	0.0745	9.8042	5
0.2561	0.0736	11.3173	0.3256	0.0749	9.9431	6
0.2282	0.0743	11.7308	0.3159	0.0752	9.2086	7

Framework versions

• Transformers 4.32.0.dev0
• TensorFlow 2.12.0
• Tokenizers 0.13.3