metadata
base_model: openai/whisper-tiny
license: apache-2.0
metrics:
- wer
tags:
- generated_from_trainer
model-index:
- name: whisper-tinyf
results: []
whisper-tinyf
This model is a fine-tuned version of openai/whisper-tiny on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.0000
- Wer: 0.0
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: 1
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 100
- training_steps: 1000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| No log | 100.0 | 100 | 0.0001 | 0.0 |
| No log | 200.0 | 200 | 0.0000 | 0.0 |
| No log | 300.0 | 300 | 0.0000 | 0.0 |
| No log | 400.0 | 400 | 0.0000 | 0.0 |
| 0.006 | 500.0 | 500 | 0.0000 | 0.0 |
| 0.006 | 600.0 | 600 | 0.0000 | 0.0 |
| 0.006 | 700.0 | 700 | 0.0000 | 0.0 |
| 0.006 | 800.0 | 800 | 0.0000 | 0.0 |
| 0.006 | 900.0 | 900 | 0.0000 | 0.0 |
| 0.0 | 1000.0 | 1000 | 0.0000 | 0.0 |
Framework versions
- Transformers 4.40.2
- Pytorch 2.2.1+cu121
- Datasets 2.19.1
- Tokenizers 0.19.1