speecht5_tts_commonvoice_official_polish_4000_5k-2

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: 0.3859

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: Use 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: 100
  • training_steps: 1000
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss
0.5246 0.4728 100 0.4775
0.4798 0.9456 200 0.4254
0.4618 1.4161 300 0.4084
0.4505 1.8889 400 0.4000
0.4383 2.3593 500 0.3932
0.4374 2.8322 600 0.3913
0.4249 3.3026 700 0.3881
0.4382 3.7754 800 0.3882
0.4222 4.2459 900 0.3856
0.4248 4.7187 1000 0.3859

Framework versions

  • Transformers 4.48.0.dev0
  • Pytorch 2.5.1+cu121
  • Datasets 3.1.0
  • Tokenizers 0.21.0
Downloads last month
30
Safetensors
Model size
144M params
Tensor type
F32
·
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Model tree for Dawid511/speecht5_tts_commonvoice_official_polish_4000_5k-2

Finetuned
(857)
this model