metadata
language:
- en
license: apache-2.0
base_model: openai/whisper-tiny
tags:
- generated_from_trainer
metrics:
- wer
model-index:
- name: whisper_tiny-mini2
results: []
whisper_tiny-mini2
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: 0.1008
- Wer: 5.5449
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: 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: 1000
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss | Wer |
---|---|---|---|---|
2.0118 | 1.49 | 100 | 1.8148 | 15.4876 |
0.3222 | 2.99 | 200 | 0.3016 | 8.0306 |
0.0315 | 4.48 | 300 | 0.1206 | 4.7801 |
0.0098 | 5.97 | 400 | 0.1097 | 4.7801 |
0.0025 | 7.46 | 500 | 0.1024 | 5.3537 |
0.0017 | 8.96 | 600 | 0.1044 | 5.1625 |
0.0011 | 10.45 | 700 | 0.1033 | 4.7801 |
0.001 | 11.94 | 800 | 0.1013 | 5.5449 |
0.0008 | 13.43 | 900 | 0.1009 | 5.3537 |
0.0008 | 14.93 | 1000 | 0.1008 | 5.5449 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.1.0+cu118
- Datasets 2.15.0
- Tokenizers 0.15.0