metadata
language:
- en
license: apache-2.0
base_model: openai/whisper-medium
tags:
- hf-asr-leaderboard
- generated_from_trainer
metrics:
- wer
model-index:
- name: Whisper Base EN
results: []
Whisper Base EN
This model is a fine-tuned version of openai/whisper-medium on the ADLINK dataset. It achieves the following results on the evaluation set:
- Loss: 0.0000
- Wer: 33.6364
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: 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: 500
- training_steps: 1000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.4926 | 25.0 | 100 | 0.1674 | 57.5758 |
| 0.0002 | 50.0 | 200 | 0.0001 | 3.9394 |
| 0.0001 | 75.0 | 300 | 0.0001 | 5.1515 |
| 0.0001 | 100.0 | 400 | 0.0001 | 10.3030 |
| 0.0001 | 125.0 | 500 | 0.0001 | 11.5152 |
| 0.0 | 150.0 | 600 | 0.0000 | 28.4848 |
| 0.0 | 175.0 | 700 | 0.0000 | 30.0 |
| 0.0 | 200.0 | 800 | 0.0000 | 29.0909 |
| 0.0 | 225.0 | 900 | 0.0000 | 33.6364 |
| 0.0 | 250.0 | 1000 | 0.0000 | 33.6364 |
Framework versions
- Transformers 4.41.1
- Pytorch 2.3.0a0+ebedce2
- Datasets 2.19.1
- Tokenizers 0.19.1