metadata
license: mit
base_model: microsoft/speecht5_tts
tags:
- generated_from_trainer
datasets:
- voxpopuli
model-index:
- name: speecht5_finetuned_voxpopuli_nl
results: []
speecht5_finetuned_voxpopuli_nl
This model is a fine-tuned version of microsoft/speecht5_tts on the voxpopuli dataset. It achieves the following results on the evaluation set:
- Loss: 0.4817
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: 4
- 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: 500
- training_steps: 1000
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.7685 | 0.43 | 100 | 0.6720 |
| 0.7072 | 0.86 | 200 | 0.6247 |
| 0.6094 | 1.29 | 300 | 0.5385 |
| 0.5648 | 1.72 | 400 | 0.5098 |
| 0.5602 | 2.15 | 500 | 0.5003 |
| 0.5337 | 2.58 | 600 | 0.4931 |
| 0.5357 | 3.01 | 700 | 0.4881 |
| 0.5315 | 3.44 | 800 | 0.4841 |
| 0.5248 | 3.87 | 900 | 0.4828 |
| 0.5281 | 4.3 | 1000 | 0.4817 |
Framework versions
- Transformers 4.31.0.dev0
- Pytorch 2.0.1+cu118
- Datasets 2.13.1
- Tokenizers 0.13.3