ESP_Model

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: 0.4896

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: Adam with betas=(0.9,0.999) and epsilon=1e-08
• 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
0.5777	11.1111	100	0.5336
0.5385	22.2222	200	0.5073
0.5054	33.3333	300	0.5005
0.4953	44.4444	400	0.4983
0.4714	55.5556	500	0.4939
0.4662	66.6667	600	0.4946
0.4563	77.7778	700	0.4981
0.4451	88.8889	800	0.4921
0.4412	100.0	900	0.4946
0.4384	111.1111	1000	0.4868
0.4308	122.2222	1100	0.4946
0.4226	133.3333	1200	0.4924
0.4186	144.4444	1300	0.4936
0.4205	155.5556	1400	0.4891
0.418	166.6667	1500	0.4896

Framework versions

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