yao_tts_yao

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: 0.0582

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
0.0801	11.6297	1000	0.0667
0.0717	23.2566	2000	0.0627
0.0657	34.8863	3000	0.0605
0.0623	46.5131	4000	0.0603
0.0589	58.1399	5000	0.0583
0.0555	69.7697	6000	0.0595
0.0542	81.3965	7000	0.0589
0.0532	93.0233	8000	0.0584
0.0508	104.6531	9000	0.0584
0.0497	116.2799	10000	0.0583
0.0476	127.9096	11000	0.0580
0.0468	139.5364	12000	0.0588
0.0474	151.1633	13000	0.0588
0.0488	162.7930	14000	0.0590
0.0482	174.4198	15000	0.0582

Framework versions

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