uz_2301_3.2_tts

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: 0.4348

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: 0.0003
• train_batch_size: 16
• eval_batch_size: 8
• seed: 42
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine_with_restarts
• lr_scheduler_warmup_steps: 500
• training_steps: 5000

Training results

Training Loss	Epoch	Step	Validation Loss
0.5459	5.0	500	0.4792
0.4878	10.0	1000	0.4627
0.4536	15.0	1500	0.4321
0.4266	20.0	2000	0.4219
0.4083	25.0	2500	0.4137
0.3864	30.0	3000	0.4259
0.3667	35.0	3500	0.4216
0.3489	40.0	4000	0.4299
0.3447	45.0	4500	0.4332
0.3445	50.0	5000	0.4348

Framework versions

• Transformers 4.48.3
• Pytorch 2.5.1+cu121
• Datasets 3.2.0
• Tokenizers 0.21.0