igbotts_refined

This model is a fine-tuned version of microsoft/speecht5_tts on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 0.4884

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: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
  • 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
0.594 3.6538 500 0.5317
0.5516 7.3012 1000 0.5032
0.534 10.9550 1500 0.4977
0.5165 14.6024 2000 0.4930
0.507 18.2498 2500 0.4911
0.5144 21.9036 3000 0.4851
0.5025 25.5510 3500 0.4885
0.5 29.1983 4000 0.4884

Framework versions

  • Transformers 4.51.3
  • Pytorch 2.6.0+cu124
  • Datasets 3.6.0
  • Tokenizers 0.21.1
Downloads last month
15
Safetensors
Model size
144M params
Tensor type
F32
ยท
Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐Ÿ™‹ Ask for provider support

Model tree for kosinebolisa/igbotts_refined

Finetuned
(1212)
this model