<!-- 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. -->
This model is a fine-tuned version of facebook/wav2vec2-xls-r-300m on the MOZILLA-FOUNDATION/COMMON_VOICE_7_0 - HI dataset. It achieves the following results on the evaluation set:
- Loss: 0.4484
- Wer: 1.0145
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: 7.5e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 2000
- num_epochs: 50.0
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 5.1844 | 3.4 | 500 | 5.2015 | 0.9999 |
| 3.3962 | 6.8 | 1000 | 3.4017 | 1.0002 |
| 2.5433 | 10.2 | 1500 | 1.6884 | 1.0222 |
| 1.5099 | 13.6 | 2000 | 0.7929 | 1.0188 |
| 1.2685 | 17.01 | 2500 | 0.6122 | 1.0191 |
| 1.1844 | 20.41 | 3000 | 0.5434 | 1.0197 |
| 1.0945 | 23.81 | 3500 | 0.5208 | 1.0316 |
| 1.0506 | 27.21 | 4000 | 0.4941 | 1.0139 |
| 1.0199 | 30.61 | 4500 | 0.4736 | 1.0106 |
| 0.9546 | 34.01 | 5000 | 0.4664 | 1.0164 |
| 0.9388 | 37.41 | 5500 | 0.4565 | 1.0085 |
| 0.9125 | 40.81 | 6000 | 0.4636 | 1.0148 |
| 0.8733 | 44.22 | 6500 | 0.4530 | 1.0154 |
| 0.8829 | 47.62 | 7000 | 0.4494 | 1.0152 |
Framework versions
- Transformers 4.16.0.dev0
- Pytorch 1.10.1+cu102
- Datasets 1.17.1.dev0
- Tokenizers 0.11.0
