learning_rate: 5e-06
train_batch_size: 4
eval_batch_size: 4
seed: 42
optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: constant_with_warmup
lr_scheduler_warmup_steps: 50
training_steps: 500

Training Loss	Epoch	Step	Validation Loss	Global Wer
0.4455	0.0110	50	0.4738	57.0848
0.4144	0.0220	100	0.4683	57.2372
0.4547	0.0330	150	0.4403	51.4982
0.372	0.0440	200	0.4035	52.8695
0.3483	0.0550	250	0.4415	48.7557
0.3796	0.0660	300	0.4004	43.3723
0.3855	0.0770	350	0.3922	43.0675
0.4376	0.0880	400	0.3859	46.3179
0.3751	0.0990	450	0.3829	42.1026
0.3226	0.1100	500	0.3794	40.0711

dianavdavidson/wh_l_v3_t_iv_indic_voices_51704_trial

wh_l_v3_t_iv_indic_voices_51704_trial