metadata
tags:
- generated_from_trainer
datasets:
- librispeech_asr
model-index:
- name: ''
results: []
This model was trained from scratch on the librispeech_asr dataset. It achieves the following results on the evaluation set:
- Loss: 3.9280
- Wer: 0.9561
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: 3e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- num_epochs: 3.0
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 5.3825 | 0.28 | 500 | 5.5398 | 0.9602 |
| 4.6164 | 0.56 | 1000 | 4.8871 | 0.9781 |
| 4.5587 | 0.84 | 1500 | 4.6683 | 1.0010 |
| 4.3428 | 1.12 | 2000 | 4.4745 | 1.6443 |
| 4.2545 | 1.4 | 2500 | 4.2854 | 1.3727 |
| 4.0861 | 1.68 | 3000 | 4.1451 | 1.3667 |
| 4.0882 | 1.96 | 3500 | 4.0432 | 1.2540 |
| 3.5985 | 2.24 | 4000 | 4.0155 | 1.0706 |
| 3.7731 | 2.52 | 4500 | 3.9633 | 1.0627 |
| 3.6984 | 2.8 | 5000 | 3.9280 | 0.9561 |
Framework versions
- Transformers 4.17.0.dev0
- Pytorch 1.10.2+cu113
- Datasets 1.18.3
- Tokenizers 0.11.0