speecht5_female

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: 0.3825

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-06
• train_batch_size: 4
• eval_batch_size: 2
• seed: 42
• gradient_accumulation_steps: 8
• total_train_batch_size: 32
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• training_steps: 10000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
0.5271	7.14	1000	0.4677
0.4785	14.27	2000	0.4240
0.4562	21.41	3000	0.4070
0.4528	28.55	4000	0.3984
0.4338	35.68	5000	0.3925
0.4387	42.82	6000	0.3890
0.4329	49.96	7000	0.3867
0.4318	57.09	8000	0.3856
0.4274	64.23	9000	0.3842
0.4287	71.36	10000	0.3825

Framework versions

• Transformers 4.39.0.dev0
• Pytorch 2.2.1
• Datasets 2.18.0
• Tokenizers 0.15.2