train_from_raw_cv12_true_suppress_syl__0020

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: 0.2180
• Train Accuracy: 0.0346
• Train Wermet: 12.3704
• Validation Loss: 0.6228
• Validation Accuracy: 0.0232
• Validation Wermet: 17.3325
• Epoch: 19

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': 2e-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
5.0429	0.0114	9.0587	3.8750	0.0118	5.0498	0
4.4427	0.0127	4.5412	3.5504	0.0125	1.0435	1
4.0670	0.0138	1.4433	3.2998	0.0131	0.8708	2
3.8001	0.0146	1.2315	3.1614	0.0134	0.9290	3
3.6184	0.0152	1.1992	3.0624	0.0137	0.8952	4
3.4663	0.0157	1.3336	3.0000	0.0139	1.1129	5
3.3304	0.0161	1.6150	2.9249	0.0142	0.9842	6
3.1968	0.0167	1.6722	2.8195	0.0145	1.0184	7
3.0193	0.0175	1.8390	2.6745	0.0151	1.0552	8
2.7798	0.0186	2.9387	2.3953	0.0161	1.5900	9
2.3558	0.0208	3.8370	1.9310	0.0178	2.0889	10
1.6793	0.0246	3.1644	1.3969	0.0199	1.8707	11
1.1330	0.0280	3.7080	1.0962	0.0212	3.6178	12
0.8349	0.0299	4.2648	0.9345	0.0218	3.0073	13
0.6496	0.0312	5.7322	0.8221	0.0223	7.7657	14
0.5173	0.0321	7.0026	0.7525	0.0226	8.9539	15
0.4172	0.0329	8.4535	0.6965	0.0229	15.4346	16
0.3370	0.0335	11.7827	0.6571	0.0231	12.8277	17
0.2722	0.0341	12.3988	0.6395	0.0232	17.4580	18
0.2180	0.0346	12.3704	0.6228	0.0232	17.3325	19

Framework versions

• Transformers 4.33.0.dev0
• TensorFlow 2.13.0
• Tokenizers 0.13.3