sinhala_check3

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: 0.6132

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: 5e-05
• train_batch_size: 1
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 8
• total_train_batch_size: 8
• optimizer: Use OptimizerNames.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: 20
• num_epochs: 25
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
0.6211	1.0	141	0.6635
0.4836	2.0	282	0.6593
0.4692	3.0	423	0.6421
0.4605	4.0	564	0.6382
0.4547	5.0	705	0.6334
0.4464	6.0	846	0.6317
0.4418	7.0	987	0.6295
0.4383	8.0	1128	0.6264
0.4339	9.0	1269	0.6286
0.4324	10.0	1410	0.6233
0.4286	11.0	1551	0.6207
0.4269	12.0	1692	0.6195
0.4246	13.0	1833	0.6191
0.42	14.0	1974	0.6258
0.4168	15.0	2115	0.6161
0.4174	16.0	2256	0.6193
0.4162	17.0	2397	0.6152
0.4132	18.0	2538	0.6167
0.4115	19.0	2679	0.6176
0.4095	20.0	2820	0.6167
0.408	21.0	2961	0.6154
0.4077	22.0	3102	0.6150
0.4055	23.0	3243	0.6144
0.4031	24.0	3384	0.6124
0.4068	24.8249	3500	0.6132

Framework versions

• Transformers 4.51.3
• Pytorch 2.6.0+cu124
• Datasets 2.14.4
• Tokenizers 0.21.1