speecht5_nyf_bible

This model is a fine-tuned version of microsoft/speecht5_tts on the arrow dataset. It achieves the following results on the evaluation set:

• Loss: 0.4393

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: 1e-05
• train_batch_size: 16
• eval_batch_size: 2
• seed: 42
• 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: 500
• training_steps: 4000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
0.5393	1.5060	500	0.4780
0.4947	3.0120	1000	0.4578
0.4738	4.5181	1500	0.4498
0.4813	6.0241	2000	0.4454
0.4755	7.5301	2500	0.4425
0.4765	9.0361	3000	0.4406
0.4641	10.5422	3500	0.4395
0.466	12.0482	4000	0.4393

Framework versions

• Transformers 4.51.3
• Pytorch 2.6.0+cu124
• Datasets 3.5.1
• Tokenizers 0.21.1