metadata
license: apache-2.0
tags:
- generated_from_trainer
metrics:
- wer
model-index:
- name: whisper-large-et-clinic
results: []
whisper-large-et-clinic
This model is a fine-tuned version of agnesluhtaru/whisper-large-et-ERR2020-v2 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.4132
- Wer: 15.7347
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: 1e-05
- train_batch_size: 2
- eval_batch_size: 1
- seed: 42
- gradient_accumulation_steps: 16
- 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: 200
- training_steps: 4000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.0297 | 4.03 | 500 | 0.3037 | 16.7147 |
| 0.0057 | 8.06 | 1000 | 0.3435 | 15.9496 |
| 0.0049 | 12.1 | 1500 | 0.3610 | 16.8737 |
| 0.0015 | 16.13 | 2000 | 0.3610 | 16.1258 |
| 0.001 | 20.16 | 2500 | 0.3868 | 16.0356 |
| 0.0011 | 24.19 | 3000 | 0.3949 | 15.8293 |
| 0.0006 | 28.22 | 3500 | 0.4028 | 15.7476 |
| 0.0008 | 32.26 | 4000 | 0.4132 | 15.7347 |
Framework versions
- Transformers 4.26.0.dev0
- Pytorch 1.12.1+rocm5.1.1
- Datasets 2.7.1.dev0
- Tokenizers 0.13.2