speecht5_finetuned_voice_dataset_bn_v_3

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: 0.5008

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-05
• 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: 125
• training_steps: 3000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
0.6046	12.2699	250	0.5646
0.5583	24.5399	500	0.5268
0.5364	36.8098	750	0.5188
0.5171	49.0798	1000	0.5087
0.5098	61.3497	1250	0.5018
0.501	73.6196	1500	0.5022
0.4984	85.8896	1750	0.4955
0.4929	98.1595	2000	0.5000
0.4933	110.4294	2250	0.4944
0.4868	122.6994	2500	0.5006
0.4805	134.9693	2750	0.4991
0.4802	147.2393	3000	0.5008

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.1+cu121
• Datasets 3.0.1
• Tokenizers 0.19.1