metadata
license: mit
base_model: microsoft/speecht5_tts
tags:
- generated_from_trainer
datasets:
- common_voice_16_0
model-index:
- name: speecht5_finetuned_commonvoice_da
results: []
speecht5_finetuned_commonvoice_da
This model is a fine-tuned version of microsoft/speecht5_tts on the common_voice_16_0 dataset. It achieves the following results on the evaluation set:
- Loss: 0.4693
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: 4
- eval_batch_size: 2
- seed: 42
- gradient_accumulation_steps: 8
- 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: 4000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.6003 | 5.13 | 500 | 0.5255 |
| 0.5378 | 10.27 | 1000 | 0.4966 |
| 0.5223 | 15.4 | 1500 | 0.4840 |
| 0.5129 | 20.54 | 2000 | 0.4760 |
| 0.5059 | 25.67 | 2500 | 0.4714 |
| 0.4942 | 30.81 | 3000 | 0.4711 |
| 0.4944 | 35.94 | 3500 | 0.4691 |
| 0.497 | 41.08 | 4000 | 0.4693 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.1.0+cu121
- Datasets 2.16.1
- Tokenizers 0.15.0