Edit model card

ceb_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: 0.3930

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
  • gradient_accumulation_steps: 2
  • total_train_batch_size: 64
  • 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
0.5525 19.6078 500 0.4686
0.4756 39.2157 1000 0.4276
0.4543 58.8235 1500 0.4116
0.4346 78.4314 2000 0.4028
0.4292 98.0392 2500 0.3997
0.4166 117.6471 3000 0.3952
0.4122 137.2549 3500 0.3957
0.4063 156.8627 4000 0.3940
0.4028 176.4706 4500 0.3951
0.3982 196.0784 5000 0.3931
0.4055 215.6863 5500 0.3946
0.4019 235.2941 6000 0.3925
0.4 254.9020 6500 0.3940
0.4046 274.5098 7000 0.3953
0.3955 294.1176 7500 0.3945
0.3944 313.7255 8000 0.3930

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