Edit model card

fil_b64_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: 0.4119

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: 64
  • 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
0.6071 21.7391 500 0.5213
0.5126 43.4783 1000 0.4507
0.4749 65.2174 1500 0.4311
0.454 86.9565 2000 0.4231
0.443 108.6957 2500 0.4173
0.4376 130.4348 3000 0.4169
0.4287 152.1739 3500 0.4133
0.4264 173.9130 4000 0.4150
0.423 195.6522 4500 0.4134
0.4223 217.3913 5000 0.4113
0.4104 239.1304 5500 0.4098
0.4192 260.8696 6000 0.4106
0.4089 282.6087 6500 0.4122
0.4146 304.3478 7000 0.4115
0.4116 326.0870 7500 0.4111
0.4097 347.8261 8000 0.4119

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