The_700_data_model

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: 0.4588

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: 100
• training_steps: 1500
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
0.5723	5.1613	100	0.5148
0.5244	10.3226	200	0.4888
0.5008	15.4839	300	0.4817
0.4873	20.6452	400	0.4742
0.4769	25.8065	500	0.4700
0.4741	30.9677	600	0.4667
0.4612	36.1290	700	0.4600
0.4522	41.2903	800	0.4593
0.4526	46.4516	900	0.4601
0.4447	51.6129	1000	0.4584
0.4489	56.7742	1100	0.4604
0.4373	61.9355	1200	0.4577
0.4332	67.0968	1300	0.4592
0.4338	72.2581	1400	0.4598
0.4328	77.4194	1500	0.4588

Framework versions

• Transformers 4.46.0.dev0
• Pytorch 2.4.1+cu121
• Datasets 3.0.1
• Tokenizers 0.20.0