metadata
license: apache-2.0
base_model: facebook/wav2vec2-large-xlsr-53
tags:
- generated_from_trainer
metrics:
- wer
model-index:
- name: Model_G_2
results: []
Model_G_2
This model is a fine-tuned version of facebook/wav2vec2-large-xlsr-53 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.7332
- Wer: 1.0098
- Cer: 0.7490
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: 0.0003
- train_batch_size: 16
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- 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: 500
- num_epochs: 30
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
|---|---|---|---|---|---|
| 3.5212 | 2.57 | 400 | 0.7741 | 1.0236 | 0.7838 |
| 0.3158 | 5.14 | 800 | 0.6119 | 1.0085 | 0.7600 |
| 0.1522 | 7.72 | 1200 | 0.6402 | 1.0215 | 0.7521 |
| 0.102 | 10.29 | 1600 | 0.6226 | 1.0134 | 0.7540 |
| 0.0752 | 12.86 | 2000 | 0.6474 | 1.0365 | 0.7501 |
| 0.0627 | 15.43 | 2400 | 0.6617 | 1.0169 | 0.7503 |
| 0.0535 | 18.01 | 2800 | 0.6818 | 1.0116 | 0.7495 |
| 0.0432 | 20.58 | 3200 | 0.7056 | 1.0125 | 0.7536 |
| 0.0383 | 23.15 | 3600 | 0.6953 | 1.0096 | 0.7448 |
| 0.0347 | 25.72 | 4000 | 0.7217 | 1.0202 | 0.7457 |
| 0.0301 | 28.3 | 4400 | 0.7332 | 1.0098 | 0.7490 |
Framework versions
- Transformers 4.31.0
- Pytorch 2.0.1+cu117
- Datasets 1.18.3
- Tokenizers 0.13.3