Hugging Face's logo

fahadqazi
/
Sindhi-TTS

Text-to-Audio
Transformers
TensorBoard
Safetensors
speecht5
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
Sindhi-TTS / README.md
fahadqazi's picture
fahadqazi
End of training
267862e verified
preview code
|
raw
history blame
4.15 kB
metadata
library_name: transformers
license: mit
base_model: Sana1207/Hindi_SpeechT5_finetuned
tags:
  - generated_from_trainer
model-index:
  - name: Sindhi-TTS
    results: []

Sindhi-TTS

This model is a fine-tuned version of Sana1207/Hindi_SpeechT5_finetuned on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 0.4887

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: 0.0001
  • train_batch_size: 16
  • eval_batch_size: 2
  • seed: 42
  • gradient_accumulation_steps: 4
  • total_train_batch_size: 64
  • 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: 5000
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss
0.5769 0.3992 100 0.5673
0.5602 0.7984 200 0.5603
0.553 1.1986 300 0.5489
0.5413 1.5978 400 0.5433
0.5361 1.9970 500 0.5336
0.5391 2.3972 600 0.5414
0.524 2.7964 700 0.5290
0.5253 3.1966 800 0.5317
0.5272 3.5958 900 0.5234
0.5258 3.9950 1000 0.5280
0.5173 4.3952 1100 0.5180
0.518 4.7944 1200 0.5207
0.5119 5.1946 1300 0.5157
0.5114 5.5938 1400 0.5169
0.5138 5.9930 1500 0.5140
0.5048 6.3932 1600 0.5122
0.5059 6.7924 1700 0.5173
0.4991 7.1926 1800 0.5057
0.5038 7.5918 1900 0.5053
0.4994 7.9910 2000 0.5071
0.4989 8.3912 2100 0.5080
0.4951 8.7904 2200 0.5099
0.4941 9.1906 2300 0.5022
0.493 9.5898 2400 0.5039
0.4915 9.9890 2500 0.5014
0.4911 10.3892 2600 0.5066
0.4861 10.7884 2700 0.4987
0.4875 11.1886 2800 0.5042
0.4892 11.5878 2900 0.4980
0.4909 11.9870 3000 0.5007
0.4886 12.3872 3100 0.4980
0.4857 12.7864 3200 0.4952
0.4868 13.1866 3300 0.4972
0.482 13.5858 3400 0.4957
0.479 13.9850 3500 0.5029
0.48 14.3852 3600 0.4954
0.4819 14.7844 3700 0.4982
0.4791 15.1846 3800 0.4936
0.4776 15.5838 3900 0.4947
0.4736 15.9830 4000 0.4930
0.4744 16.3832 4100 0.4937
0.4735 16.7824 4200 0.4895
0.4789 17.1826 4300 0.4936
0.4729 17.5818 4400 0.4920
0.4742 17.9810 4500 0.4915
0.4721 18.3812 4600 0.4887
0.4733 18.7804 4700 0.4933
0.4849 19.1806 4800 0.4879
0.4692 19.5798 4900 0.4889
0.4747 19.9790 5000 0.4887

Framework versions

  • Transformers 4.46.2
  • Pytorch 2.5.0+cu121
  • Datasets 3.1.0
  • Tokenizers 0.20.3