senga-nt-fused-v3-tts

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: 0.0904

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: 4000
• training_steps: 40000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
0.1649	9.2173	2000	0.1052
0.1553	18.4347	4000	0.1013
0.1389	27.6520	6000	0.0997
0.1356	36.8694	8000	0.0967
0.1275	46.0832	10000	0.0954
0.1208	55.3006	12000	0.0950
0.1199	64.5179	14000	0.0934
0.1138	73.7353	16000	0.0943
0.1171	82.9526	18000	0.0933
0.1065	92.1665	20000	0.0924
0.1018	101.3838	22000	0.0927
0.0986	110.6012	24000	0.0916
0.1006	119.8185	26000	0.0912
0.0968	129.0324	28000	0.0914
0.0939	138.2497	30000	0.0908
0.0941	147.4671	32000	0.0904
0.0911	156.6844	34000	0.0911
0.0913	165.9017	36000	0.0905
0.0926	175.1156	38000	0.0905
0.0916	184.3329	40000	0.0904

Framework versions

• Transformers 4.57.1
• Pytorch 2.8.0+cu128
• Datasets 4.2.0
• Tokenizers 0.22.2