metadata
base_model: openai/whisper-tiny
tags:
- generated_from_trainer
datasets:
- PolyAI/minds14
metrics:
- wer
model-index:
- name: whisper-tiny-polyai-enUS_fewer_epochs
results:
- task:
name: Automatic Speech Recognition
type: automatic-speech-recognition
dataset:
name: PolyAI/minds14
type: PolyAI/minds14
metrics:
- name: Wer
type: wer
value: 0.34946871310507677
whisper-tiny-polyai-enUS_fewer_epochs
This model is a fine-tuned version of openai/whisper-tiny on the PolyAI/minds14 dataset. It achieves the following results on the evaluation set:
- Loss: 0.6145
- Wer Ortho: 0.3800
- Wer: 0.3495
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: 5e-06
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: constant_with_warmup
- lr_scheduler_warmup_steps: 50
- training_steps: 200
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer Ortho | Wer |
|---|---|---|---|---|---|
| 2.9576 | 3.33 | 50 | 1.9424 | 0.5077 | 0.4050 |
| 0.5132 | 6.67 | 100 | 0.6382 | 0.4152 | 0.3684 |
| 0.2569 | 10.0 | 150 | 0.5925 | 0.3893 | 0.3554 |
| 0.0973 | 13.33 | 200 | 0.6145 | 0.3800 | 0.3495 |
Framework versions
- Transformers 4.36.2
- Pytorch 2.1.1+cu121
- Datasets 2.16.1
- Tokenizers 0.15.0