metadata
license: apache-2.0
tags:
- generated_from_trainer
base_model: openai/whisper-tiny
model-index:
- name: tiny_test1
results: []
tiny_test1
This model is a fine-tuned version of openai/whisper-tiny on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.0436
- Cer: 30.9792
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: 16
- 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: 500
- training_steps: 4000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Cer |
|---|---|---|---|---|
| 0.3366 | 5.59 | 1000 | 0.8486 | 32.4585 |
| 0.0591 | 11.17 | 2000 | 0.9387 | 30.4734 |
| 0.0145 | 16.76 | 3000 | 1.0169 | 30.4925 |
| 0.0086 | 22.35 | 4000 | 1.0436 | 30.9792 |
Framework versions
- Transformers 4.38.2
- Pytorch 2.0.0
- Datasets 2.18.0
- Tokenizers 0.15.2