metadata
license: apache-2.0
base_model: openai/whisper-base.en
tags:
- generated_from_trainer
metrics:
- wer
model-index:
- name: abbenedekwhisper-base.en-finetuning2-D3K
results: []
abbenedekwhisper-base.en-finetuning2-D3K
This model is a fine-tuned version of openai/whisper-base.en on the None dataset. It achieves the following results on the evaluation set:
- Loss: 6.9927
- Cer: 63.0495
- Wer: 127.1523
- Ser: 100.0
- Cer Clean: 4.3406
- Wer Clean: 7.2848
- Ser Clean: 9.6491
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-08
- train_batch_size: 16
- eval_batch_size: 64
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 5
- training_steps: 100
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Cer | Wer | Ser | Cer Clean | Wer Clean | Ser Clean |
|---|---|---|---|---|---|---|---|---|---|
| 7.8089 | 0.11 | 20 | 6.9970 | 63.0495 | 127.1523 | 100.0 | 4.3406 | 7.2848 | 9.6491 |
| 7.653 | 0.21 | 40 | 6.9963 | 63.3834 | 127.4834 | 100.0 | 4.6745 | 7.6159 | 10.5263 |
| 7.6425 | 0.32 | 60 | 6.9951 | 63.3834 | 127.4834 | 100.0 | 4.6745 | 7.6159 | 10.5263 |
| 7.6689 | 0.43 | 80 | 6.9952 | 63.0495 | 127.1523 | 100.0 | 4.3406 | 7.2848 | 9.6491 |
| 7.6972 | 0.53 | 100 | 6.9927 | 63.0495 | 127.1523 | 100.0 | 4.3406 | 7.2848 | 9.6491 |
Framework versions
- Transformers 4.39.3
- Pytorch 2.2.2+cu121
- Datasets 2.14.5
- Tokenizers 0.15.2