speecht5_finetuned_ugspeech_ak

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: 0.3654

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 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: 5000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
0.4378	1.4744	500	0.4069
0.4306	2.9488	1000	0.3975
0.4141	4.4231	1500	0.3862
0.4082	5.8975	2000	0.3851
0.4029	7.3738	2500	0.3819
0.4021	8.8481	3000	0.3795
0.4016	11.4744	3500	0.3763
0.3918	13.1131	4000	0.3745
0.3889	13.2735	4500	0.3682
0.3886	14.7479	5000	0.3654

Framework versions

• Transformers 4.46.2
• Pytorch 2.5.1+cu121
• Datasets 3.1.0
• Tokenizers 0.20.3