speecht5_finetuned_9ja

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: 0.4223

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: 1000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
0.6144	3.5714	100	0.4972
0.4638	7.1429	200	0.4306
0.4315	10.7143	300	0.4087
0.4071	14.2857	400	0.4117
0.3901	17.8571	500	0.4000
0.3808	21.4286	600	0.3989
0.3672	25.0	700	0.4052
0.3584	28.5714	800	0.4187
0.3439	32.1429	900	0.4229
0.3428	35.7143	1000	0.4223

Framework versions

• Transformers 4.47.0
• Pytorch 2.5.1+cu121
• Datasets 3.3.1
• Tokenizers 0.21.0