speecht5_finetuned_librispeech_polish_epo10_batch2_gas2

This model is a fine-tuned version of dawid511/speecht5_finetuned_librispeech_polish_epo6_batch8_gas4 on an unknown dataset. It achieves the following results on the evaluation set:

Model description

More information needed

More information needed

More information needed

The following hyperparameters were used during training:

learning_rate: 0.0001
train_batch_size: 2
eval_batch_size: 2
seed: 42
gradient_accumulation_steps: 2
total_train_batch_size: 4
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
num_epochs: 10
mixed_precision_training: Native AMP

Training Loss	Epoch	Step	Validation Loss
0.7279	0.2558	100	0.3734
0.7647	0.5115	200	0.3829
0.7646	0.7673	300	0.3793
0.7521	1.0230	400	0.3804
0.7673	1.2788	500	0.3817
0.7415	1.5345	600	0.3824
0.7721	1.7903	700	0.3960
0.7766	2.0460	800	0.3767
0.7529	2.3018	900	0.3756
0.757	2.5575	1000	0.3809
0.757	2.8133	1100	0.3808
0.746	3.0691	1200	0.3762
0.7424	3.3248	1300	0.3744
0.7409	3.5806	1400	0.3778
0.7453	3.8363	1500	0.3715
0.7409	4.0921	1600	0.3722
0.7441	4.3478	1700	0.3728
0.7304	4.6036	1800	0.3724
0.738	4.8593	1900	0.3710
0.7213	5.1151	2000	0.3730
0.7446	5.3708	2100	0.3721
0.7255	5.6266	2200	0.3684
0.7321	5.8824	2300	0.3671
0.7098	6.1381	2400	0.3673
0.7401	6.3939	2500	0.3735
0.7165	6.6496	2600	0.3679
0.714	6.9054	2700	0.3733
0.7035	7.1611	2800	0.3666
0.7089	7.4169	2900	0.3689
0.7118	7.6726	3000	0.3691
0.7064	7.9284	3100	0.3664
0.6994	8.1841	3200	0.3679
0.6958	8.4399	3300	0.3661
0.7087	8.6957	3400	0.3683
0.6968	8.9514	3500	0.3635
0.7035	9.2072	3600	0.3647
0.7045	9.4629	3700	0.3647
0.6982	9.7187	3800	0.3642
0.6996	9.9744	3900	0.3637