speecht5_finetuned_kaito_darija

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: 0.4338

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 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
0.6036	0.2790	100	0.5324
0.5641	0.5581	200	0.5118
0.5207	0.8371	300	0.4781
0.5018	1.1144	400	0.4696
0.499	1.3934	500	0.4631
0.4967	1.6725	600	0.4572
0.4791	1.9515	700	0.4503
0.4737	2.2288	800	0.4399
0.4713	2.5078	900	0.4383
0.4683	2.7869	1000	0.4338

Framework versions

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