<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
speecht5_finetuned_voxpopuli_es
This model is a fine-tuned version of KGSAGAR/speecht5_finetuned_voxpopuli_es on the voxpopuli dataset. It achieves the following results on the evaluation set:
- Loss: 0.5950
Model description
More information needed
Intended uses & limitations
(https://colab.research.google.com/drive/1NG4uTeW97wYRdzkjfWI4gONntxg9Y17S?usp=sharing)__ To enhance the model's performance, it is advisable to increase the training steps parameter and carry out further training.
Training and evaluation data
TrainOutput(global_step=25, training_loss=0.6927243232727051, metrics={'train_runtime': 8513.8366, 'train_samples_per_second': 0.094, 'train_steps_per_second': 0.003, 'total_flos': 116396101622592.0, 'train_loss': 0.6927243232727051, 'epoch': 0.11})
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 3e-05
- train_batch_size: 4
- eval_batch_size: 2
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 5
- training_steps: 25
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.7148 | 0.02 | 5 | 0.6356 |
| 0.7004 | 0.04 | 10 | 0.6357 |
| 0.6845 | 0.06 | 15 | 0.6040 |
| 0.6813 | 0.09 | 20 | 0.5962 |
| 0.6827 | 0.11 | 25 | 0.5950 |
Framework versions
- Transformers 4.32.0
- Pytorch 2.0.1+cu118
- Datasets 2.14.4
- Tokenizers 0.13.3
