metadata
license: apache-2.0
base_model: openai/whisper-tiny
tags:
- generated_from_trainer
metrics:
- wer
model-index:
- name: whisper-tiny-en
results: []
datasets:
- PolyAI/minds14
pipeline_tag: automatic-speech-recognition
whisper-tiny-en
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: 0.5603
- Wer Ortho: 0.2844
- Wer: 0.2910
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: 1
- eval_batch_size: 1
- 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: 2225
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer Ortho | Wer |
|---|---|---|---|---|---|
| 0.5396 | 1.0 | 445 | 0.4247 | 0.3387 | 0.3394 |
| 0.2289 | 2.0 | 890 | 0.4628 | 0.2961 | 0.3017 |
| 0.1448 | 3.0 | 1335 | 0.4680 | 0.2819 | 0.2869 |
| 0.0405 | 4.0 | 1780 | 0.5402 | 0.3029 | 0.3052 |
| 0.0092 | 5.0 | 2225 | 0.5603 | 0.2844 | 0.2910 |
Framework versions
- Transformers 4.44.0
- Pytorch 2.1.1+cu118
- Datasets 2.20.0
- Tokenizers 0.19.1