ceb_b64_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: 0.4050

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: 4
• total_train_batch_size: 64
• 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
0.4561	19.8020	500	0.4151
0.4179	39.6040	1000	0.3994
0.4075	59.4059	1500	0.4018
0.3981	79.2079	2000	0.4029
0.3862	99.0099	2500	0.3978
0.3726	118.8119	3000	0.3978
0.365	138.6139	3500	0.3960
0.3525	158.4158	4000	0.3969
0.3545	178.2178	4500	0.3982
0.3473	198.0198	5000	0.4039
0.3439	217.8218	5500	0.4020
0.3371	237.6238	6000	0.4044
0.3362	257.4257	6500	0.4041
0.3311	277.2277	7000	0.4022
0.3345	297.0297	7500	0.4051
0.3348	316.8317	8000	0.4050

Framework versions

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