xwhisper-kh-tiny
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: 0.2922
- Wer: 81.9342
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: constant
- lr_scheduler_warmup_steps: 1000
- num_epochs: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 1.0917 | 1.0 | 1087 | 0.5887 | 97.9640 |
| 0.4597 | 2.0 | 2174 | 0.4018 | 89.2179 |
| 0.3283 | 3.0 | 3261 | 0.3462 | 87.5905 |
| 0.2635 | 4.0 | 4348 | 0.3151 | 86.8306 |
| 0.2197 | 5.0 | 5435 | 0.2969 | 84.9237 |
| 0.1855 | 6.0 | 6522 | 0.2936 | 81.8983 |
| 0.1574 | 7.0 | 7609 | 0.2898 | 83.6691 |
| 0.1336 | 8.0 | 8696 | 0.2922 | 81.9342 |
Framework versions
- Transformers 4.45.2
- Pytorch 2.4.0
- Datasets 3.1.0
- Tokenizers 0.20.3
- Downloads last month
- 144
Inference Providers
NEW
This model is not currently available via any of the supported third-party Inference Providers, and
the model is not deployed on the HF Inference API.
Model tree for PhanithLIM/xwhisper-kh-tiny
Base model
openai/whisper-tiny