metadata
license: mit
base_model: microsoft/speecht5_tts
tags:
- generated_from_trainer
model-index:
- name: zlm_b64_le5_s8000
results: []
zlm_b64_le5_s8000
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: 0.3630
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: 32
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 2000
- training_steps: 8500
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.4525 | 0.4188 | 500 | 0.4043 |
| 0.4541 | 0.8375 | 1000 | 0.3992 |
| 0.4355 | 1.2563 | 1500 | 0.3946 |
| 0.4315 | 1.6750 | 2000 | 0.3966 |
| 0.4329 | 2.0938 | 2500 | 0.3881 |
| 0.4235 | 2.5126 | 3000 | 0.3829 |
| 0.4179 | 2.9313 | 3500 | 0.3775 |
| 0.4116 | 3.3501 | 4000 | 0.3739 |
| 0.4107 | 3.7688 | 4500 | 0.3721 |
| 0.4029 | 4.1876 | 5000 | 0.3693 |
| 0.409 | 4.6064 | 5500 | 0.3680 |
| 0.4061 | 5.0251 | 6000 | 0.3662 |
| 0.403 | 5.4439 | 6500 | 0.3654 |
| 0.3958 | 5.8626 | 7000 | 0.3630 |
| 0.3952 | 6.2814 | 7500 | 0.3635 |
| 0.3971 | 6.7002 | 8000 | 0.3627 |
| 0.4004 | 7.1189 | 8500 | 0.3630 |
Framework versions
- Transformers 4.41.0.dev0
- Pytorch 2.2.1+cu121
- Datasets 2.19.0
- Tokenizers 0.19.1