frieren_english_tenganai2

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: 0.5361

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: 4
• eval_batch_size: 2
• seed: 42
• gradient_accumulation_steps: 8
• total_train_batch_size: 32
• optimizer: Use OptimizerNames.ADAMW_TORCH 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: 100
• training_steps: 1500
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
0.513	10.0	100	0.5191
0.4472	20.0	200	0.5186
0.4157	30.0	300	0.5035
0.3997	40.0	400	0.5354
0.3902	50.0	500	0.5078
0.376	60.0	600	0.5026
0.3725	70.0	700	0.5102
0.357	80.0	800	0.5242
0.3566	90.0	900	0.5162
0.3477	100.0	1000	0.5373
0.3441	110.0	1100	0.5388
0.3396	120.0	1200	0.5324
0.3354	130.0	1300	0.5211
0.3298	140.0	1400	0.5389
0.3349	150.0	1500	0.5361

Framework versions

• Transformers 4.52.2
• Pytorch 2.6.0+cu124
• Datasets 3.6.0
• Tokenizers 0.21.1