speecht5_finetuned_id_covost

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: 0.4778

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: 2
• seed: 42
• gradient_accumulation_steps: 8
• total_train_batch_size: 64
• 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: 100
• training_steps: 1000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
0.6435	0.5966	100	0.5643
0.5774	1.1969	200	0.5396
0.5628	1.7934	300	0.5233
0.5429	2.3937	400	0.5003
0.527	2.9903	500	0.4960
0.5242	3.5906	600	0.4893
0.5161	4.1909	700	0.4880
0.5088	4.7875	800	0.4843
0.5034	5.3878	900	0.4776
0.4961	5.9843	1000	0.4778

Framework versions

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