<!-- 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. -->
wav2vec2-burak-new-300-v2-6
This model is a fine-tuned version of facebook/wav2vec2-xls-r-300m on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.3074
- Wer: 0.2340
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: 16
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- num_epochs: 151
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 6.3136 | 9.61 | 500 | 3.1262 | 1.0 |
| 1.8247 | 19.23 | 1000 | 0.4049 | 0.5065 |
| 0.5387 | 28.83 | 1500 | 0.2828 | 0.3462 |
| 0.3713 | 38.45 | 2000 | 0.2761 | 0.3125 |
| 0.293 | 48.08 | 2500 | 0.2872 | 0.3001 |
| 0.2436 | 57.68 | 3000 | 0.2912 | 0.2904 |
| 0.2116 | 67.3 | 3500 | 0.2910 | 0.2725 |
| 0.1859 | 76.91 | 4000 | 0.2937 | 0.2533 |
| 0.1731 | 86.53 | 4500 | 0.2985 | 0.2485 |
| 0.1569 | 96.15 | 5000 | 0.3022 | 0.2409 |
| 0.1471 | 105.76 | 5500 | 0.3070 | 0.2374 |
| 0.1385 | 115.38 | 6000 | 0.2954 | 0.2429 |
| 0.1289 | 124.99 | 6500 | 0.3016 | 0.2361 |
| 0.1268 | 134.61 | 7000 | 0.3000 | 0.2368 |
| 0.12 | 144.23 | 7500 | 0.3074 | 0.2340 |
Framework versions
- Transformers 4.24.0
- Pytorch 1.12.1+cu113
- Datasets 2.6.1
- Tokenizers 0.13.2
