metadata
license: mit
base_model: microsoft/speecht5_tts
tags:
- generated_from_trainer
model-index:
- name: malay_norm_checkpoint
results: []
malay_norm_checkpoint
This model is a fine-tuned version of microsoft/speecht5_tts on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.4922
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: 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: 500
- training_steps: 8000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.8434 | 6.81 | 1000 | 0.7457 |
| 0.6769 | 13.63 | 2000 | 0.6074 |
| 0.6142 | 20.44 | 3000 | 0.5630 |
| 0.5925 | 27.26 | 4000 | 0.5253 |
| 0.5579 | 34.07 | 5000 | 0.5163 |
| 0.5459 | 40.89 | 6000 | 0.4985 |
| 0.5428 | 47.7 | 7000 | 0.4960 |
| 0.5443 | 54.51 | 8000 | 0.4922 |
Framework versions
- Transformers 4.40.0.dev0
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2