<!-- 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. -->
STT_Model_11
This model is a fine-tuned version of facebook/wav2vec2-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.1127
- Wer: 0.428
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: 8
- 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: 50
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| No log | 4.88 | 200 | 4.0668 | 1.0 |
| No log | 9.76 | 400 | 3.4212 | 1.0 |
| 7.5598 | 14.63 | 600 | 2.2886 | 1.0 |
| 7.5598 | 19.51 | 800 | 0.5655 | 0.69 |
| 1.0223 | 24.39 | 1000 | 0.3529 | 0.534 |
| 1.0223 | 29.27 | 1200 | 0.2115 | 0.504 |
| 1.0223 | 34.15 | 1400 | 0.2567 | 0.464 |
| 0.1478 | 39.02 | 1600 | 0.3085 | 0.456 |
| 0.1478 | 43.9 | 1800 | 0.1537 | 0.444 |
| 0.034 | 48.78 | 2000 | 0.1127 | 0.428 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.13.1+cu116
- Datasets 2.10.1
- Tokenizers 0.13.2
