yao_tts_kde

This model is a fine-tuned version of microsoft/speecht5_tts on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.0592

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: 8
• eval_batch_size: 8
• seed: 3407
• gradient_accumulation_steps: 4
• total_train_batch_size: 32
• 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: cosine
• lr_scheduler_warmup_steps: 1500
• training_steps: 15000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
0.08	11.6297	1000	0.0676
0.0717	23.2566	2000	0.0629
0.0651	34.8863	3000	0.0616
0.0631	46.5131	4000	0.0599
0.0592	58.1399	5000	0.0597
0.0558	69.7697	6000	0.0594
0.0546	81.3965	7000	0.0610
0.0534	93.0233	8000	0.0594
0.0512	104.6531	9000	0.0594
0.0502	116.2799	10000	0.0593
0.0479	127.9096	11000	0.0597
0.047	139.5364	12000	0.0598
0.0478	151.1633	13000	0.0596
0.0492	162.7930	14000	0.0600
0.0484	174.4198	15000	0.0592

Framework versions

• Transformers 4.57.1
• Pytorch 2.8.0+cu128
• Datasets 3.6.0
• Tokenizers 0.22.2