fil_b32_le5_s4000 / README.md
mikhail-panzo's picture
End of training
e260f9f
|
raw
history blame
1.82 kB
metadata
license: mit
base_model: microsoft/speecht5_tts
tags:
  - generated_from_trainer
model-index:
  - name: fil_b32_le5_s4000
    results: []

fil_b32_le5_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: 0.4139

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
0.6246 11.1111 500 0.5376
0.5034 22.2222 1000 0.4566
0.4807 33.3333 1500 0.4360
0.4582 44.4444 2000 0.4245
0.4541 55.5556 2500 0.4213
0.4409 66.6667 3000 0.4158
0.4469 77.7778 3500 0.4153
0.4337 88.8889 4000 0.4139

Framework versions

  • Transformers 4.41.0.dev0
  • Pytorch 2.2.1+cu121
  • Datasets 2.19.0
  • Tokenizers 0.19.1