fil_b32_le4_s8000

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: 0.4087

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: 16
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 2
• 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: 2000
• training_steps: 8000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
0.498	10.9890	500	0.4393
0.448	21.9780	1000	0.4195
0.4411	32.9670	1500	0.4205
0.4347	43.9560	2000	0.4253
0.4173	54.9451	2500	0.4151
0.4012	65.9341	3000	0.4118
0.4023	76.9231	3500	0.4092
0.3873	87.9121	4000	0.4116
0.381	98.9011	4500	0.4089
0.3804	109.8901	5000	0.4093
0.3724	120.8791	5500	0.4066
0.3665	131.8681	6000	0.4092
0.3635	142.8571	6500	0.4099
0.3562	153.8462	7000	0.4075
0.3581	164.8352	7500	0.4097
0.3461	175.8242	8000	0.4087

Framework versions

• Transformers 4.41.0.dev0
• Pytorch 2.2.1+cu121
• Datasets 2.19.0
• Tokenizers 0.19.1