Edit model card

fil_b32_le5_s8000

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: 0.4039

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: 32
  • eval_batch_size: 8
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_steps: 2000
  • training_steps: 8000
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss
0.632 11.1111 500 0.5323
0.519 22.2222 1000 0.4494
0.4816 33.3333 1500 0.4291
0.481 44.4444 2000 0.4211
0.4459 55.5556 2500 0.4139
0.4484 66.6667 3000 0.4114
0.4317 77.7778 3500 0.4081
0.4301 88.8889 4000 0.4076
0.4274 100.0 4500 0.4059
0.4323 111.1111 5000 0.4062
0.4189 122.2222 5500 0.4045
0.4272 133.3333 6000 0.4059
0.4219 144.4444 6500 0.4058
0.4125 155.5556 7000 0.4049
0.42 166.6667 7500 0.4046
0.4145 177.7778 8000 0.4039

Framework versions

  • Transformers 4.41.0.dev0
  • Pytorch 2.2.1+cu121
  • Datasets 2.19.0
  • Tokenizers 0.19.1
Downloads last month
0
Safetensors
Model size
144M params
Tensor type
F32
·

Finetuned from