whisper-small-pashto-run9-freshbase

This model is a fine-tuned version of openai/whisper-small 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:

Training Loss	Epoch	Step	Validation Loss	Wer
1.8567	0.0137	100	1.2021	68.7395
1.3236	0.0275	200	0.9091	57.7591
1.0413	0.0412	300	0.8083	55.4622
0.9023	0.0549	400	0.7765	53.4454
0.8436	0.0686	500	0.7404	54.4164
0.8362	0.0824	600	0.7322	52.4370
0.7782	0.0961	700	0.7142	49.2810
0.7893	0.1098	800	0.7027	50.4202
0.6939	0.1235	900	0.6737	48.9076
0.6584	0.1373	1000	0.6934	47.3950
0.6082	0.1510	1100	0.6610	47.7871
0.596	0.1647	1200	0.6639	47.3389
0.5677	0.1784	1300	0.6685	47.7311
0.5502	0.1922	1400	0.6715	45.9010
0.4807	0.2059	1500	0.6769	44.2390
0.4931	0.2196	1600	0.6951	43.4174
0.535	0.2334	1700	0.6742	43.4734
0.4746	0.2471	1800	0.6990	44.2764
0.4759	0.2608	1900	0.6873	44.4818
0.4199	0.2745	2000	0.7065	43.2493
0.3889	0.2883	2100	0.6965	43.9029
0.4014	0.3020	2200	0.6694	43.0626
0.3971	0.3157	2300	0.6652	44.8366
0.434	0.3294	2400	0.6956	45.5649
0.3495	0.3432	2500	0.6922	43.3240
0.3652	0.3569	2600	0.6998	44.3511
0.5335	0.3706	2700	0.6737	42.9318
0.4788	0.3844	2800	0.6575	42.4090
0.4456	0.3981	2900	0.6615	41.9234
0.4576	0.4118	3000	0.6576	42.5957
0.4298	0.4255	3100	0.6735	44.8366
0.4062	0.4393	3200	0.6792	45.0980
0.4101	0.4530	3300	0.6823	42.1102
0.3945	0.4667	3400	0.6683	41.2138
0.3935	0.4804	3500	0.6951	43.7722
0.3789	0.4942	3600	0.6772	42.5397
0.3809	0.5079	3700	0.7004	43.0812
0.349	0.5216	3800	0.7114	43.6041
0.3945	0.5353	3900	0.7049	42.2782