metadata
license: apache-2.0
base_model: openai/whisper-tiny
tags:
- generated_from_trainer
metrics:
- wer
model-index:
- name: whisper-tiny-atcosim
results: []
whisper-tiny-atcosim
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: 0.0718
- Wer: 72.8885
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: 128
- eval_batch_size: 128
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 10
- num_epochs: 100
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.2141 | 8.33 | 500 | 0.0623 | 15.6696 |
| 0.0021 | 16.67 | 1000 | 0.0636 | 34.9463 |
| 0.0007 | 25.0 | 1500 | 0.0676 | 47.4023 |
| 0.0003 | 33.33 | 2000 | 0.0666 | 57.0383 |
| 0.0002 | 41.67 | 2500 | 0.0677 | 63.1413 |
| 0.0002 | 50.0 | 3000 | 0.0686 | 64.6879 |
| 0.0001 | 58.33 | 3500 | 0.0694 | 68.1608 |
| 0.0001 | 66.67 | 4000 | 0.0701 | 70.0639 |
| 0.0001 | 75.0 | 4500 | 0.0708 | 71.2956 |
| 0.0001 | 83.33 | 5000 | 0.0713 | 72.0411 |
| 0.0001 | 91.67 | 5500 | 0.0716 | 72.8839 |
| 0.0001 | 100.0 | 6000 | 0.0718 | 72.8885 |
Framework versions
- Transformers 4.36.2
- Pytorch 2.1.2
- Datasets 2.15.0
- Tokenizers 0.15.0