<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
combined-onTorgoSentenceModel-TestedSpeakerM01
This model is a fine-tuned version of alexziweiwang/torgo-sentences-TestedSpeakerM01 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.3823
- Wer: 0.8064
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: 0.0001
- train_batch_size: 4
- 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: 1000
- num_epochs: 30
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 28.7173 | 0.35 | 500 | 3.0048 | 0.9974 |
| 2.5941 | 0.69 | 1000 | 2.1227 | 1.1821 |
| 1.9528 | 1.04 | 1500 | 1.8921 | 1.2128 |
| 1.5166 | 1.39 | 2000 | 1.3280 | 1.1795 |
| 1.3394 | 1.74 | 2500 | 1.3925 | 1.0923 |
| 1.2284 | 2.08 | 3000 | 1.4628 | 1.1064 |
| 1.0785 | 2.43 | 3500 | 1.3823 | 1.0923 |
| 1.0343 | 2.78 | 4000 | 1.2196 | 1.0923 |
| 0.9979 | 3.12 | 4500 | 1.4429 | 1.0038 |
| 0.8963 | 3.47 | 5000 | 1.3360 | 0.9731 |
| 0.9169 | 3.82 | 5500 | 1.4309 | 0.9949 |
| 0.7811 | 4.17 | 6000 | 1.4731 | 0.9974 |
| 0.7811 | 4.51 | 6500 | 1.5859 | 0.9808 |
| 0.7818 | 4.86 | 7000 | 1.3996 | 0.9923 |
| 0.6952 | 5.21 | 7500 | 1.2495 | 1.0167 |
| 0.6708 | 5.56 | 8000 | 1.7330 | 1.0103 |
| 0.6724 | 5.9 | 8500 | 1.0807 | 0.9731 |
| 0.6321 | 6.25 | 9000 | 1.4906 | 0.9833 |
| 0.6564 | 6.6 | 9500 | 1.2854 | 0.9218 |
| 0.6099 | 6.94 | 10000 | 1.4289 | 0.9321 |
| 0.5643 | 7.29 | 10500 | 1.5796 | 0.9449 |
| 0.5612 | 7.64 | 11000 | 1.3855 | 0.9013 |
| 0.5746 | 7.99 | 11500 | 0.9218 | 0.8936 |
| 0.5198 | 8.33 | 12000 | 1.0982 | 0.9436 |
| 0.5754 | 8.68 | 12500 | 1.1686 | 0.9449 |
| 0.5102 | 9.03 | 13000 | 1.1525 | 0.9321 |
| 0.4721 | 9.38 | 13500 | 1.2355 | 0.8551 |
| 0.4715 | 9.72 | 14000 | 1.2203 | 0.8897 |
| 0.4961 | 10.07 | 14500 | 1.3226 | 0.8821 |
| 0.4256 | 10.42 | 15000 | 1.4078 | 0.8795 |
| 0.449 | 10.76 | 15500 | 1.1023 | 0.8731 |
| 0.4141 | 11.11 | 16000 | 1.0989 | 0.9026 |
| 0.4325 | 11.46 | 16500 | 1.3532 | 0.8936 |
| 0.4034 | 11.81 | 17000 | 1.0583 | 0.8705 |
| 0.4027 | 12.15 | 17500 | 1.0765 | 0.8692 |
| 0.4031 | 12.5 | 18000 | 1.4424 | 0.8679 |
| 0.4129 | 12.85 | 18500 | 1.1504 | 0.8654 |
| 0.4291 | 13.19 | 19000 | 1.3790 | 0.8692 |
| 0.3512 | 13.54 | 19500 | 1.2891 | 0.9077 |
| 0.3499 | 13.89 | 20000 | 1.3068 | 0.8577 |
| 0.3331 | 14.24 | 20500 | 1.5288 | 0.8833 |
| 0.3626 | 14.58 | 21000 | 1.1218 | 0.9090 |
| 0.495 | 14.93 | 21500 | 1.2152 | 0.8641 |
| 0.3129 | 15.28 | 22000 | 1.4101 | 0.8974 |
| 0.2945 | 15.62 | 22500 | 1.2965 | 0.8615 |
| 0.3335 | 15.97 | 23000 | 1.3362 | 0.8282 |
| 0.2902 | 16.32 | 23500 | 1.2603 | 0.8679 |
| 0.294 | 16.67 | 24000 | 1.4634 | 0.8603 |
| 0.2561 | 17.01 | 24500 | 1.0413 | 0.8333 |
| 0.2654 | 17.36 | 25000 | 1.5734 | 0.8679 |
| 0.2724 | 17.71 | 25500 | 1.6416 | 0.8756 |
| 0.2581 | 18.06 | 26000 | 1.3018 | 0.8551 |
| 0.2329 | 18.4 | 26500 | 1.5288 | 0.8654 |
| 0.2484 | 18.75 | 27000 | 1.4365 | 0.8410 |
| 0.2578 | 19.1 | 27500 | 1.4501 | 0.8564 |
| 0.2237 | 19.44 | 28000 | 1.3333 | 0.8436 |
| 0.2274 | 19.79 | 28500 | 1.3387 | 0.8256 |
| 0.2253 | 20.14 | 29000 | 1.4898 | 0.8551 |
| 0.2189 | 20.49 | 29500 | 1.6225 | 0.8692 |
| 0.2006 | 20.83 | 30000 | 1.6085 | 0.8551 |
| 0.2263 | 21.18 | 30500 | 1.5435 | 0.85 |
| 0.2032 | 21.53 | 31000 | 1.3926 | 0.8372 |
| 0.2128 | 21.88 | 31500 | 1.4497 | 0.8397 |
| 0.1776 | 22.22 | 32000 | 1.5413 | 0.8372 |
| 0.1962 | 22.57 | 32500 | 1.4021 | 0.8410 |
| 0.1823 | 22.92 | 33000 | 1.5397 | 0.8410 |
| 0.1634 | 23.26 | 33500 | 1.5256 | 0.8141 |
| 0.1971 | 23.61 | 34000 | 1.4673 | 0.8308 |
| 0.1926 | 23.96 | 34500 | 1.4941 | 0.8167 |
| 0.1695 | 24.31 | 35000 | 1.5740 | 0.7949 |
| 0.1555 | 24.65 | 35500 | 1.5857 | 0.8269 |
| 0.1714 | 25.0 | 36000 | 1.4801 | 0.8333 |
| 0.1393 | 25.35 | 36500 | 1.5619 | 0.8359 |
| 0.1364 | 25.69 | 37000 | 1.5414 | 0.8385 |
| 0.1571 | 26.04 | 37500 | 1.4345 | 0.8205 |
| 0.1605 | 26.39 | 38000 | 1.4427 | 0.8192 |
| 0.1473 | 26.74 | 38500 | 1.4327 | 0.8179 |
| 0.1575 | 27.08 | 39000 | 1.3873 | 0.8090 |
| 0.1454 | 27.43 | 39500 | 1.3101 | 0.8103 |
| 0.1197 | 27.78 | 40000 | 1.3747 | 0.8038 |
| 0.1301 | 28.12 | 40500 | 1.3558 | 0.7962 |
| 0.1416 | 28.47 | 41000 | 1.2969 | 0.7962 |
| 0.1642 | 28.82 | 41500 | 1.3601 | 0.8 |
| 0.1234 | 29.17 | 42000 | 1.3910 | 0.8051 |
| 0.1262 | 29.51 | 42500 | 1.3941 | 0.8077 |
| 0.1408 | 29.86 | 43000 | 1.3823 | 0.8064 |
Framework versions
- Transformers 4.23.1
- Pytorch 1.12.1+cu113
- Datasets 1.18.3
- Tokenizers 0.13.2
