speecht5_finetuned_voxpopuli_ro

This model is a fine-tuned version of microsoft/speecht5_tts on the voxpopuli dataset. It achieves the following results on the evaluation set:

• Loss: 0.4358

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: 8
• eval_batch_size: 4
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 32
• 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: 200
• training_steps: 1000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
0.587	3.1221	100	0.4987
0.5147	6.2443	200	0.4616
0.4926	9.3664	300	0.4524
0.4711	12.4885	400	0.4428
0.4643	15.6107	500	0.4415
0.4537	18.7328	600	0.4396
0.446	21.8550	700	0.4379
0.4419	24.9771	800	0.4367
0.4412	28.1221	900	0.4338
0.4365	31.2443	1000	0.4358

Framework versions

• Transformers 4.48.3
• Pytorch 2.5.1+cu124
• Datasets 3.3.2
• Tokenizers 0.21.0