zlm_b128
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: 2.7907
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: 0.001
- train_batch_size: 16
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 128
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 2000
- training_steps: 4000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.4638 | 0.8377 | 500 | 0.4116 |
| 0.4543 | 1.6754 | 1000 | 0.3974 |
| 2.8163 | 2.5131 | 1500 | 2.8015 |
| 2.8164 | 3.3508 | 2000 | 2.7967 |
| 2.8045 | 4.1885 | 2500 | 2.7965 |
| 2.8146 | 5.0262 | 3000 | 2.7904 |
| 2.8105 | 5.8639 | 3500 | 2.7910 |
| 2.8075 | 6.7016 | 4000 | 2.7907 |
Framework versions
- Transformers 4.41.0.dev0
- Pytorch 2.2.1+cu121
- Datasets 2.19.0
- Tokenizers 0.19.1
- Downloads last month
- 1