speecht5-ngiemboon

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: 0.5662

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: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 16
• 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: 10
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
1.4586	0.7375	500	0.7303
1.2905	1.4749	1000	0.6574
1.2284	2.2124	1500	0.6263
1.1975	2.9499	2000	0.6020
1.1748	3.6873	2500	0.5899
1.1447	4.4248	3000	0.5836
1.1359	5.1622	3500	0.5814
1.1298	5.8997	4000	0.5734
1.1204	6.6372	4500	0.5719
1.1073	7.3746	5000	0.5696
1.1296	8.1121	5500	0.5688
1.1152	8.8496	6000	0.5667
1.1067	9.5870	6500	0.5662

Framework versions

• Transformers 5.0.0
• Pytorch 2.10.0+cu128
• Datasets 2.18.0
• Tokenizers 0.22.2