|
--- |
|
license: mit |
|
base_model: microsoft/speecht5_tts |
|
tags: |
|
- generated_from_trainer |
|
model-index: |
|
- name: zlm_b32_le5_s12000 |
|
results: [] |
|
--- |
|
|
|
<!-- 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. --> |
|
|
|
# zlm_b32_le5_s12000 |
|
|
|
This model is a fine-tuned version of [microsoft/speecht5_tts](https://huggingface.co/microsoft/speecht5_tts) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.3707 |
|
|
|
## 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: 16 |
|
- eval_batch_size: 8 |
|
- seed: 42 |
|
- gradient_accumulation_steps: 2 |
|
- 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: 2000 |
|
- training_steps: 12000 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:------:|:-----:|:---------------:| |
|
| 0.7211 | 0.2094 | 500 | 0.6148 | |
|
| 0.6059 | 0.4188 | 1000 | 0.5140 | |
|
| 0.5347 | 0.6283 | 1500 | 0.4725 | |
|
| 0.4888 | 0.8377 | 2000 | 0.4612 | |
|
| 0.4923 | 1.0471 | 2500 | 0.4283 | |
|
| 0.466 | 1.2565 | 3000 | 0.4163 | |
|
| 0.4535 | 1.4660 | 3500 | 0.4090 | |
|
| 0.4442 | 1.6754 | 4000 | 0.4009 | |
|
| 0.4423 | 1.8848 | 4500 | 0.3955 | |
|
| 0.4539 | 2.0942 | 5000 | 0.3916 | |
|
| 0.4416 | 2.3037 | 5500 | 0.3870 | |
|
| 0.4306 | 2.5131 | 6000 | 0.3856 | |
|
| 0.4242 | 2.7225 | 6500 | 0.3819 | |
|
| 0.426 | 2.9319 | 7000 | 0.3814 | |
|
| 0.4105 | 3.1414 | 7500 | 0.3787 | |
|
| 0.4077 | 3.3508 | 8000 | 0.3750 | |
|
| 0.4106 | 3.5602 | 8500 | 0.3748 | |
|
| 0.4228 | 3.7696 | 9000 | 0.3728 | |
|
| 0.4101 | 3.9791 | 9500 | 0.3719 | |
|
| 0.4209 | 4.1885 | 10000 | 0.3707 | |
|
| 0.4091 | 4.3979 | 10500 | 0.3712 | |
|
| 0.4061 | 4.6073 | 11000 | 0.3715 | |
|
| 0.4169 | 4.8168 | 11500 | 0.3700 | |
|
| 0.4088 | 5.0262 | 12000 | 0.3707 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.41.0.dev0 |
|
- Pytorch 2.2.1+cu121 |
|
- Datasets 2.19.0 |
|
- Tokenizers 0.19.1 |
|
|
