senga-nt-canon-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: 0.0929

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
0.168	9.2166	2000	0.1071
0.1609	18.4332	4000	0.1057
0.1443	27.6498	6000	0.1004
0.1381	36.8664	8000	0.0978
0.1305	46.0829	10000	0.0976
0.1196	55.2995	12000	0.0975
0.1176	64.5161	14000	0.0962
0.1154	73.7327	16000	0.0951
0.1153	82.9493	18000	0.0962
0.1083	92.1659	20000	0.0948
0.1076	101.3825	22000	0.0951
0.1028	110.5991	24000	0.0944
0.0992	119.8157	26000	0.0940
0.0944	129.0323	28000	0.0949
0.0956	138.2488	30000	0.0941
0.092	147.4654	32000	0.0931
0.094	156.6820	34000	0.0938
0.0908	165.8986	36000	0.0932
0.0883	175.1152	38000	0.0938
0.0903	184.3318	40000	0.0929

Framework versions

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