<!-- This model card has been generated automatically according to the information Keras had access to. You should probably proofread and complete it, then remove this comment. -->
whisper_syl_noforce_nostart__0045
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: 0.1322
- Train Accuracy: 0.0333
- Train Wermet: 0.0508
- Validation Loss: 0.7727
- Validation Accuracy: 0.0207
- Validation Wermet: 0.3008
- Epoch: 44
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 |
---|---|---|---|---|---|---|
5.6298 | 0.0091 | 1.6176 | 4.3084 | 0.0092 | 1.0203 | 0 |
4.9271 | 0.0098 | 0.8937 | 4.1324 | 0.0099 | 0.9075 | 1 |
4.6878 | 0.0106 | 0.8360 | 3.9151 | 0.0102 | 0.9003 | 2 |
4.4454 | 0.0113 | 0.8275 | 3.7558 | 0.0106 | 0.8730 | 3 |
4.2497 | 0.0119 | 0.8211 | 3.6019 | 0.0110 | 0.8640 | 4 |
4.0917 | 0.0123 | 0.8067 | 3.5363 | 0.0111 | 0.8512 | 5 |
3.9616 | 0.0127 | 0.7864 | 3.4492 | 0.0113 | 0.8432 | 6 |
3.8575 | 0.0130 | 0.7742 | 3.3963 | 0.0113 | 0.8414 | 7 |
3.7605 | 0.0133 | 0.7580 | 3.3430 | 0.0115 | 0.8197 | 8 |
3.6756 | 0.0136 | 0.7447 | 3.2872 | 0.0117 | 0.8071 | 9 |
3.6021 | 0.0138 | 0.7370 | 3.2828 | 0.0117 | 0.8165 | 10 |
3.5237 | 0.0140 | 0.7218 | 3.2439 | 0.0118 | 0.8088 | 11 |
3.4558 | 0.0143 | 0.7105 | 3.2063 | 0.0120 | 0.7890 | 12 |
3.3853 | 0.0145 | 0.6993 | 3.1702 | 0.0120 | 0.8035 | 13 |
3.3101 | 0.0148 | 0.6870 | 3.1144 | 0.0123 | 0.7605 | 14 |
3.2314 | 0.0152 | 0.6719 | 3.0522 | 0.0125 | 0.7481 | 15 |
3.1430 | 0.0155 | 0.6575 | 2.9911 | 0.0127 | 0.7378 | 16 |
3.0392 | 0.0160 | 0.6369 | 2.9249 | 0.0129 | 0.7357 | 17 |
2.9134 | 0.0166 | 0.6148 | 2.7883 | 0.0134 | 0.6909 | 18 |
2.7583 | 0.0173 | 0.5911 | 2.6383 | 0.0139 | 0.6695 | 19 |
2.5761 | 0.0182 | 0.5649 | 2.4519 | 0.0144 | 0.6360 | 20 |
2.3668 | 0.0192 | 0.5329 | 2.2887 | 0.0150 | 0.6022 | 21 |
2.1435 | 0.0203 | 0.4947 | 2.0979 | 0.0156 | 0.5742 | 22 |
1.9309 | 0.0214 | 0.4597 | 1.9288 | 0.0162 | 0.5377 | 23 |
1.6980 | 0.0226 | 0.4192 | 1.7768 | 0.0167 | 0.5094 | 24 |
1.4712 | 0.0239 | 0.3799 | 1.6030 | 0.0173 | 0.4825 | 25 |
1.2644 | 0.0252 | 0.3419 | 1.4490 | 0.0179 | 0.4463 | 26 |
1.0965 | 0.0262 | 0.3082 | 1.3237 | 0.0184 | 0.4211 | 27 |
0.9565 | 0.0271 | 0.2781 | 1.2294 | 0.0187 | 0.3952 | 28 |
0.8423 | 0.0278 | 0.2511 | 1.1475 | 0.0191 | 0.3752 | 29 |
0.7467 | 0.0285 | 0.2281 | 1.0891 | 0.0193 | 0.3629 | 30 |
0.6625 | 0.0291 | 0.2055 | 1.0262 | 0.0195 | 0.3458 | 31 |
0.5908 | 0.0296 | 0.1855 | 0.9831 | 0.0197 | 0.3344 | 32 |
0.5255 | 0.0301 | 0.1666 | 0.9412 | 0.0199 | 0.3206 | 33 |
0.4710 | 0.0305 | 0.1499 | 0.9086 | 0.0200 | 0.3132 | 34 |
0.4179 | 0.0309 | 0.1333 | 0.8889 | 0.0201 | 0.3064 | 35 |
0.3722 | 0.0312 | 0.1178 | 0.8580 | 0.0203 | 0.2966 | 36 |
0.3310 | 0.0316 | 0.1040 | 0.8327 | 0.0204 | 0.2932 | 37 |
0.2943 | 0.0319 | 0.0927 | 0.8268 | 0.0204 | 0.2896 | 38 |
0.2598 | 0.0322 | 0.0818 | 0.8072 | 0.0205 | 0.2843 | 39 |
0.2291 | 0.0324 | 0.0713 | 0.7944 | 0.0205 | 0.2797 | 40 |
0.2006 | 0.0327 | 0.0627 | 0.7884 | 0.0206 | 0.2802 | 41 |
0.1752 | 0.0329 | 0.0557 | 0.7887 | 0.0206 | 0.2994 | 42 |
0.1523 | 0.0331 | 0.0553 | 0.7845 | 0.0206 | 0.2854 | 43 |
0.1322 | 0.0333 | 0.0508 | 0.7727 | 0.0207 | 0.3008 | 44 |
Framework versions
- Transformers 4.33.0.dev0
- TensorFlow 2.13.0
- Tokenizers 0.13.3