speecht5_finetuned_binisha

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: 0.4253

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: Use 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: 100
• training_steps: 1500
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
0.648	7.9901	100	0.5922
0.5721	15.9901	200	0.5445
0.5337	23.9901	300	0.5103
0.5057	31.9901	400	0.5052
0.4894	39.9901	500	0.4869
0.4765	47.9901	600	0.4804
0.4577	55.9901	700	0.4770
0.4462	63.9901	800	0.4561
0.4275	71.9901	900	0.4445
0.4143	79.9901	1000	0.4388
0.4044	87.9901	1100	0.4363
0.3929	95.9901	1200	0.4299
0.3922	103.9901	1300	0.4276
0.3915	111.9901	1400	0.4262
0.3877	119.9901	1500	0.4253

Framework versions

• Transformers 4.46.2
• Pytorch 2.5.1+cu121
• Datasets 3.1.0
• Tokenizers 0.20.3