learning_rate: 5e-05
train_batch_size: 1
eval_batch_size: 8
seed: 42
gradient_accumulation_steps: 8
total_train_batch_size: 8
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: 10
num_epochs: 20
mixed_precision_training: Native AMP

Training Loss	Epoch	Step	Validation Loss
0.6302	1.0	42	0.5935
0.5339	2.0	84	0.5820
0.5221	3.0	126	0.5702
0.5002	4.0	168	0.5617
0.4858	5.0	210	0.5527
0.4825	6.0	252	0.5489
0.4723	7.0	294	0.5467
0.4656	8.0	336	0.5362
0.4531	9.0	378	0.5436
0.4494	10.0	420	0.5356
0.4448	11.0	462	0.5295
0.4393	12.0	504	0.5225
0.4398	13.0	546	0.5220
0.4318	14.0	588	0.5232
0.4269	15.0	630	0.5157
0.4299	16.0	672	0.5146
0.4246	17.0	714	0.5142
0.4229	18.0	756	0.5137
0.4178	19.0	798	0.5153
0.4191	19.5350	820	0.5140

kavinda123321/speecht5_finetuned_english_mahinda_aug

speecht5_finetuned_english_mahinda_aug