metadata
license: apache-2.0
base_model: openai/whisper-tiny
tags:
- generated_from_trainer
metrics:
- wer
model-index:
- name: whisper-tinyfinacial
results: []
whisper-tinyfinacial
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: 4.1540
- Wer: 154.4944
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: 600
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| No log | 25.0 | 50 | 1.4261 | 66.8539 |
| No log | 50.0 | 100 | 1.3916 | 86.5169 |
| No log | 75.0 | 150 | 1.6553 | 165.1685 |
| No log | 100.0 | 200 | 2.6574 | 134.8315 |
| No log | 125.0 | 250 | 2.7460 | 142.6966 |
| No log | 150.0 | 300 | 3.4242 | 157.3034 |
| No log | 175.0 | 350 | 3.7021 | 165.7303 |
| No log | 200.0 | 400 | 3.9109 | 168.5393 |
| No log | 225.0 | 450 | 4.0157 | 198.8764 |
| 3.7169 | 250.0 | 500 | 4.1466 | 164.6067 |
| 3.7169 | 275.0 | 550 | 4.1483 | 152.8090 |
| 3.7169 | 300.0 | 600 | 4.1540 | 154.4944 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.3.0+cu121
- Datasets 2.20.0
- Tokenizers 0.19.1