Urdu Tts Phonemes Finetuned

--- library_name: transformers tags: - generated_from_trainer model-index: - name: urdu-tts-phonemes-finetuned results: [] --- # urdu-tts-phonemes-finetuned This model was trained from scratch on an unknown dataset. It achieves the following results on the evaluation set: - Loss: 0.8902 ## 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: 6 - eval_batch_size: 2 - seed: 42 - distributed_type: multi-GPU - num_devices: 2 - gradient_accumulation_steps: 4 - total_train_batch_size: 48 - total_eval_batch_size: 4 - optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED 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: 500 - num_epochs: 70 - mixed_precision_training: Native AMP ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:-------:|:-----:|:---------------:| | 5.4748 | 1.5163 | 500 | 1.0756 | | 4.4286 | 3.0304 | 1000 | 1.0000 | | 4.2899 | 4.5467 | 1500 | 0.9708 | | 4.1616 | 6.0607 | 2000 | 0.9536 | | 4.0716 | 7.5771 | 2500 | 0.9420 | | 4.0121 | 9.0911 | 3000 | 0.9242 | | 3.9637 | 10.6074 | 3500 | 0.9217 | | 3.9152 | 12.1215 | 4000 | 0.9091 | | 3.8967 | 13.6378 | 4500 | 0.9034 | | 3.8794 | 15.1519 | 5000 | 0.9066 | | 3.8580 | 16.6682 | 5500 | 0.9018 | | 3.8195 | 18.1822 | 6000 | 0.8976 | | 3.8034 | 19.6986 | 6500 | 0.8946 | | 3.7870 | 21.2126 | 7000 | 0.8929 | | 3.7691 | 22.7289 | 7500 | 0.8952 | | 3.7517 | 24.2430 | 8000 | 0.8890 | | 3.7299 | 25.7593 | 8500 | 0.8941 | | 3.7293 | 27.2733 | 9000 | 0.8908 | | 3.7309 | 28.7897 | 9500 | 0.8911 | | 3.7051 | 30.3037 | 10000 | 0.8860 | | 3.6962 | 31.8200 | 10500 | 0.8879 | | 3.6794 | 33.3341 | 11000 | 0.8842 | | 3.6740 | 34.8504 | 11500 | 0.8866 | | 3.6693 | 36.3645 | 12000 | 0.8834 | | 3.6793 | 37.8808 | 12500 | 0.8885 | | 3.6572 | 39.3948 | 13000 | 0.8844 | | 3.6636 | 40.9112 | 13500 | 0.8826 | | 3.6410 | 42.4252 | 14000 | 0.8840 | | 3.6616 | 43.9415 | 14500 | 0.8921 | | 3.6408 | 45.4556 | 15000 | 0.8882 | | 3.6513 | 46.9719 | 15500 | 0.8869 | | 3.6223 | 48.4860 | 16000 | 0.8887 | | 3.6251 | 50.0 | 16500 | 0.8921 | | 3.6284 | 51.5163 | 17000 | 0.8865 | | 3.6264 | 53.0304 | 17500 | 0.8910 | | 3.6112 | 54.5467 | 18000 | 0.8881 | | 3.6109 | 56.0607 | 18500 | 0.8929 | | 3.6175 | 57.5771 | 19000 | 0.8859 | | 3.6266 | 59.0911 | 19500 | 0.8897 | | 3.6035 | 60.6074 | 20000 | 0.8870 | | 3.5990 | 62.1215 | 20500 | 0.8916 | | 3.6005 | 63.6378 | 21000 | 0.8894 | | 3.6143 | 65.1519 | 21500 | 0.8857 | | 3.6044 | 66.6682 | 22000 | 0.8916 | | 3.6021 | 68.1822 | 22500 | 0.8911 | | 3.6110 | 69.6986 | 23000 | 0.8902 | ### Framework versions - Transformers 5.0.0 - Pytorch 2.10.0+cu128 - Datasets 4.8.3 - Tokenizers 0.22.2