metadata
base_model: openai/whisper-tiny
license: apache-2.0
metrics:
- wer
tags:
- generated_from_trainer
model-index:
- name: whisper-tinyfinacialKI500
results: []
whisper-tinyfinacialKI500
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: 0.5350
- Wer: 50.5618
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: 1.35e-05
- train_batch_size: 8
- 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: 100
- training_steps: 400
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| No log | 0.7937 | 50 | 0.8964 | 62.9213 |
| No log | 1.5873 | 100 | 0.6468 | 56.7416 |
| No log | 2.3810 | 150 | 0.5310 | 48.3146 |
| No log | 3.1746 | 200 | 0.4966 | 50.5618 |
| No log | 3.9683 | 250 | 0.5187 | 47.1910 |
| No log | 4.7619 | 300 | 0.5419 | 48.8764 |
| No log | 5.5556 | 350 | 0.5410 | 55.6180 |
| No log | 6.3492 | 400 | 0.5350 | 50.5618 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.3.0+cu121
- Datasets 2.20.0
- Tokenizers 0.19.1