hubert-base-ls960-speech-commands-v1-t_70
This model is a fine-tuned version of facebook/hubert-base-ls960 on the speech_commands dataset. It achieves the following results on the evaluation set:
- Loss: 1.6949
- Accuracy: 0.7914
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-05
- train_batch_size: 36
- eval_batch_size: 36
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 144
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 3
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 2.0981 | 1.0 | 275 | 2.5151 | 0.6987 |
| 1.1717 | 2.0 | 550 | 1.8235 | 0.7738 |
| 1.0304 | 2.9900 | 822 | 1.6949 | 0.7914 |
Framework versions
- Transformers 4.51.2
- Pytorch 2.6.0+cu126
- Datasets 3.5.0
- Tokenizers 0.21.1