You can deploy this model and start earning money today!

IshaanHugs/
wav2vec2-adam

Name: wav2vec2-adam
Availability: InStock

safetensors

wav2vec2

generated_from_trainer

facebook/wav2vec2-xls-r-1b

finetune

apache-2.0

Views

No views yet

wav2vec2-adam

This model is a fine-tuned version of facebook/wav2vec2-xls-r-1b on the None dataset. It achieves the following results on the evaluation set:

Loss: 568.1923
Wer: 0.4049
Cer: 0.1661

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: 8
eval_batch_size: 8
seed: 42
gradient_accumulation_steps: 2
total_train_batch_size: 16
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_ratio: 0.1
num_epochs: 15
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer	Cer
1936.1745	0.17	200	1877.9899	1.0	1.0
1826.8428	0.34	400	1617.3900	0.9999	0.9988
989.1127	0.52	600	765.3218	0.8003	0.2914
851.7201	0.69	800	628.5829	0.6985	0.2497
723.3757	0.86	1000	616.5229	0.6419	0.2340
760.4359	1.03	1200	564.5872	0.6281	0.2214
643.8634	1.21	1400	548.7975	0.6051	0.2212
643.329	1.38	1600	538.1201	0.5615	0.2077
589.8902	1.55	1800	550.2911	0.5749	0.2108
621.258	1.72	2000	500.6337	0.5573	0.2052
595.7741	1.89	2200	496.0975	0.5566	0.2034
607.9014	2.07	2400	498.4079	0.5428	0.2006
557.1155	2.24	2600	477.4254	0.5249	0.1940
584.153	2.41	2800	531.1208	0.5067	0.1915
534.6832	2.58	3000	575.7189	0.5612	0.2152
504.149	2.76	3200	563.9553	0.5486	0.2009
482.5778	2.93	3400	632.0757	0.5864	0.2268
533.353	3.1	3600	598.1741	0.5388	0.2039
509.3027	3.27	3800	678.9108	0.5858	0.2302
504.9937	3.44	4000	571.2620	0.5323	0.1998
478.2493	3.62	4200	502.2411	0.4825	0.1879
527.6547	3.79	4400	699.0823	0.6150	0.2397
518.2816	3.96	4600	496.8146	0.4685	0.1815
434.0348	4.13	4800	453.2502	0.4551	0.1782
479.1774	4.3	5000	464.5193	0.4519	0.1771
491.2223	4.48	5200	475.8923	0.4655	0.1799
424.2042	4.65	5400	529.0171	0.4611	0.1790
447.7984	4.82	5600	470.2910	0.4526	0.1782
419.982	4.99	5800	537.6948	0.4666	0.1804
465.5625	5.17	6000	483.3431	0.4678	0.1840
463.9381	5.34	6200	452.2645	0.4455	0.1767
470.612	5.51	6400	470.7729	0.4489	0.1767
424.7324	5.68	6600	512.8937	0.4515	0.1768
458.3429	5.85	6800	559.7302	0.4632	0.1790
412.2542	6.03	7000	512.7188	0.4273	0.1727
432.4705	6.2	7200	547.5737	0.4406	0.1736
449.5592	6.37	7400	521.1343	0.4456	0.1746
409.4003	6.54	7600	529.0558	0.4513	0.1775
403.7129	6.72	7800	478.3857	0.4264	0.1698
410.4987	6.89	8000	507.4272	0.4393	0.1752
414.295	7.06	8200	537.4016	0.4356	0.1712
391.8523	7.23	8400	689.75	0.4787	0.1870
405.5996	7.4	8600	553.1865	0.4264	0.1696
395.3369	7.58	8800	455.0266	0.4132	0.1664
400.0372	7.75	9000	538.5027	0.4403	0.1769
366.8186	7.92	9200	509.5075	0.4093	0.1667
407.5628	8.09	9400	489.1333	0.4310	0.1700
402.5129	8.27	9600	535.4504	0.4199	0.1699
378.6691	8.44	9800	526.0916	0.4237	0.1705
361.7635	8.61	10000	533.3074	0.4123	0.1687
375.3004	8.78	10200	477.7126	0.4060	0.1647
342.6438	8.95	10400	550.2237	0.4191	0.1690
332.2053	9.13	10600	581.2722	0.4181	0.1678
358.5789	9.3	10800	507.9084	0.4214	0.1689
362.3012	9.47	11000	503.8822	0.4069	0.1655
382.7161	9.64	11200	489.9423	0.3992	0.1640
333.7467	9.81	11400	515.5997	0.4100	0.1656
378.7749	9.99	11600	500.2072	0.4078	0.1656
310.0922	10.16	11800	592.5539	0.4024	0.1639
320.5362	10.33	12000	611.4968	0.3975	0.1629
364.4849	10.5	12200	553.2431	0.3993	0.1644
314.1099	10.68	12400	575.7576	0.4154	0.1665
356.3336	10.85	12600	548.5637	0.4060	0.1644
337.4982	11.02	12800	508.9512	0.3992	0.1630
319.0365	11.19	13000	495.1051	0.3966	0.1640
339.525	11.36	13200	552.6434	0.4020	0.1638
298.221	11.54	13400	538.9012	0.4150	0.1657
386.3358	11.71	13600	536.7173	0.4021	0.1648
331.7429	11.88	13800	518.9872	0.4070	0.1643
334.6416	12.05	14000	539.6923	0.4044	0.1645
317.3125	12.23	14200	528.7343	0.4014	0.1632
321.3217	12.4	14400	545.8629	0.3953	0.1630
324.9768	12.57	14600	559.4356	0.4006	0.1643
297.2214	12.74	14800	567.3050	0.4018	0.1649
321.2393	12.91	15000	561.3262	0.4050	0.1666
245.4859	13.09	15200	578.2697	0.4014	0.1644
302.7325	13.26	15400	599.7056	0.3948	0.1630
303.934	13.43	15600	587.3956	0.4000	0.1661
351.8048	13.6	15800	574.0975	0.3990	0.1650
295.436	13.78	16000	577.4233	0.3994	0.1645
313.0659	13.95	16200	580.9484	0.4004	0.1647
268.2741	14.12	16400	592.4994	0.4050	0.1659
325.2033	14.29	16600	582.0380	0.4046	0.1654
262.8658	14.46	16800	572.5929	0.4061	0.1665
302.6039	14.64	17000	571.0538	0.4066	0.1662
286.299	14.81	17200	568.9926	0.4061	0.1665
253.382	14.98	17400	568.1923	0.4049	0.1661

Framework versions

Transformers 4.38.2
Pytorch 2.2.0+cu118
Datasets 2.16.0
Tokenizers 0.15.2

IshaanHugs/wav2vec2-adam

wav2vec2-adam

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

IshaanHugs/
wav2vec2-adam