metadata
license: mit
base_model: microsoft/speecht5_tts
tags:
- generated_from_trainer
model-index:
- name: zlm_b32_le4_s4000
results: []
zlm_b32_le4_s4000
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: 0.3362
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: 4000
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
0.3715 | 0.2094 | 500 | 0.3407 |
0.3764 | 0.4188 | 1000 | 0.3403 |
0.3665 | 0.6283 | 1500 | 0.3418 |
0.3579 | 0.8377 | 2000 | 0.3401 |
0.3674 | 1.0471 | 2500 | 0.3386 |
0.3604 | 1.2565 | 3000 | 0.3372 |
0.3614 | 1.4660 | 3500 | 0.3368 |
0.3671 | 1.6754 | 4000 | 0.3362 |
Framework versions
- Transformers 4.41.0.dev0
- Pytorch 2.2.1+cu121
- Datasets 2.19.0
- Tokenizers 0.19.1