metadata
license: apache-2.0
tags:
- generated_from_trainer
datasets:
- common_voice
model-index:
- name: ''
results: []
This model is a fine-tuned version of facebook/wav2vec2-xls-r-1b on the common_voice dataset. It achieves the following results on the evaluation set:
- Loss: 0.6925
- Wer: 1.0517
- Cer: 0.1923
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: 32
- 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: 2000
- num_epochs: 50.0
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
|---|---|---|---|---|---|
| 1.9321 | 3.14 | 1000 | 1.0116 | 0.9823 | 0.2635 |
| 2.0934 | 6.29 | 2000 | 1.1241 | 1.0222 | 0.2932 |
| 2.0389 | 9.43 | 3000 | 1.2067 | 1.1325 | 0.3345 |
| 1.9569 | 12.58 | 4000 | 0.9818 | 1.0090 | 0.2657 |
| 1.8409 | 15.72 | 5000 | 1.0382 | 1.6480 | 0.3741 |
| 1.7449 | 18.87 | 6000 | 0.9962 | 1.6268 | 0.3454 |
| 1.7349 | 22.01 | 7000 | 0.9560 | 0.9850 | 0.2597 |
| 1.6857 | 25.16 | 8000 | 0.8722 | 0.9669 | 0.2325 |
| 1.5637 | 28.3 | 9000 | 0.7636 | 1.8071 | 0.3422 |
| 1.5088 | 31.45 | 10000 | 0.7290 | 1.0398 | 0.2085 |
| 1.4298 | 34.59 | 11000 | 0.7576 | 1.0166 | 0.2104 |
| 1.3716 | 37.74 | 12000 | 0.7046 | 1.1275 | 0.2138 |
| 1.3185 | 40.88 | 13000 | 0.7011 | 1.1696 | 0.2179 |
| 1.28 | 44.03 | 14000 | 0.6754 | 1.1316 | 0.2024 |
| 1.2368 | 47.17 | 15000 | 0.6925 | 1.0517 | 0.1923 |
Framework versions
- Transformers 4.17.0.dev0
- Pytorch 1.10.2+cu102
- Datasets 1.18.2.dev0
- Tokenizers 0.11.0