You can deploy this model and start earning money today!

IshaanHugs/
wav2vec2-hiddendrop2

Name: wav2vec2-hiddendrop2
Availability: InStock

safetensors

wav2vec2

generated_from_trainer

facebook/wav2vec2-xls-r-1b

finetune

apache-2.0

Views

No views yet

wav2vec2-hiddendrop2

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: 490.5553
Wer: 0.3913
Cer: 0.1610

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
1929.0708	0.17	200	1880.8464	1.0	1.0
1838.0028	0.34	400	1738.6960	1.0	1.0000
1103.8241	0.52	600	829.5133	0.8617	0.3098
930.7859	0.69	800	729.6690	0.7399	0.2633
794.2839	0.86	1000	674.1293	0.6899	0.2528
822.4573	1.03	1200	600.8782	0.6415	0.2303
693.8613	1.21	1400	600.8663	0.6424	0.2319
692.2782	1.38	1600	538.8715	0.6033	0.2180
636.1471	1.55	1800	517.0187	0.5884	0.2174
654.8398	1.72	2000	511.7276	0.5740	0.2093
624.9046	1.89	2200	517.5521	0.5625	0.2053
640.1055	2.07	2400	505.9084	0.5494	0.2008
583.7511	2.24	2600	500.1803	0.5416	0.1983
606.5029	2.41	2800	473.5118	0.5107	0.1933
561.5286	2.58	3000	477.6519	0.5266	0.1997
529.8504	2.76	3200	520.7839	0.5236	0.1980
505.6423	2.93	3400	492.6431	0.4973	0.1870
557.1387	3.1	3600	557.4595	0.5047	0.1931
536.9029	3.27	3800	610.7289	0.5056	0.1928
533.7711	3.44	4000	529.0969	0.4997	0.1942
498.5374	3.62	4200	525.0269	0.4926	0.1873
550.378	3.79	4400	550.3901	0.5045	0.1945
546.544	3.96	4600	562.3098	0.5140	0.1911
455.4148	4.13	4800	463.8059	0.4784	0.1841
499.0308	4.3	5000	501.0483	0.4747	0.1831
510.2767	4.48	5200	489.3989	0.4754	0.1818
443.9621	4.65	5400	491.4554	0.4599	0.1791
478.5215	4.82	5600	461.2336	0.4586	0.1792
446.3521	4.99	5800	541.8758	0.4667	0.1818
485.5593	5.17	6000	471.2174	0.4672	0.1820
489.5416	5.34	6200	490.2437	0.4689	0.1839
495.3557	5.51	6400	550.8417	0.4994	0.1949
440.0705	5.68	6600	547.6086	0.4602	0.1783
480.3493	5.85	6800	525.2051	0.4511	0.1775
435.8765	6.03	7000	523.2603	0.4595	0.1828
446.0985	6.2	7200	490.5027	0.4462	0.1739
475.9563	6.37	7400	605.2362	0.5258	0.2048
423.1882	6.54	7600	564.1501	0.4850	0.1881
425.9348	6.72	7800	452.1814	0.4391	0.1738
430.0744	6.89	8000	472.8405	0.4345	0.1735
434.0012	7.06	8200	585.6984	0.4596	0.1794
414.9007	7.23	8400	584.1519	0.4467	0.1759
417.2396	7.4	8600	533.2339	0.4541	0.1786
413.6659	7.58	8800	490.7388	0.4487	0.1751
415.661	7.75	9000	479.2579	0.4197	0.1714
380.4528	7.92	9200	509.2206	0.4403	0.1736
436.1994	8.09	9400	528.6705	0.4499	0.1744
422.7479	8.27	9600	549.8844	0.4463	0.1745
402.4668	8.44	9800	505.9262	0.4287	0.1709
383.8194	8.61	10000	500.6133	0.4355	0.1708
398.1734	8.78	10200	527.6191	0.4278	0.1697
372.1209	8.95	10400	533.9014	0.4503	0.1768
355.3869	9.13	10600	509.8195	0.4180	0.1666
378.6731	9.3	10800	543.1016	0.4623	0.1787
384.0029	9.47	11000	508.8887	0.4137	0.1661
404.5413	9.64	11200	518.2016	0.4266	0.1696
354.3923	9.81	11400	594.0906	0.4318	0.1722
405.2743	9.99	11600	497.8559	0.4264	0.1691
335.2736	10.16	11800	491.9792	0.4097	0.1647
338.5644	10.33	12000	526.3049	0.4240	0.1695
386.549	10.5	12200	487.0261	0.4080	0.1657
335.8394	10.68	12400	508.1725	0.4131	0.1641
377.0482	10.85	12600	506.3954	0.3981	0.1623
356.9405	11.02	12800	492.8109	0.4033	0.1627
339.2195	11.19	13000	518.5338	0.4195	0.1670
359.9845	11.36	13200	504.9150	0.4178	0.1672
320.5116	11.54	13400	478.3988	0.4238	0.1672
408.086	11.71	13600	458.1666	0.4063	0.1639
353.8871	11.88	13800	466.5937	0.3966	0.1614
357.7566	12.05	14000	505.6631	0.4053	0.1628
339.0464	12.23	14200	491.0547	0.4034	0.1644
341.2426	12.4	14400	501.2332	0.3959	0.1621
345.5909	12.57	14600	494.9123	0.4112	0.1666
319.4277	12.74	14800	500.2198	0.4072	0.1658
345.6838	12.91	15000	526.2410	0.4074	0.1650
268.5453	13.09	15200	498.0937	0.4035	0.1634
330.4404	13.26	15400	490.4071	0.4026	0.1629
325.3512	13.43	15600	505.7834	0.3985	0.1637
377.6949	13.6	15800	507.7114	0.3918	0.1616
323.2166	13.78	16000	491.5162	0.3927	0.1616
336.217	13.95	16200	488.9820	0.3935	0.1622
291.1413	14.12	16400	505.9708	0.3979	0.1623
347.4715	14.29	16600	505.3162	0.3995	0.1623
284.7323	14.46	16800	496.2574	0.3971	0.1619
325.0236	14.64	17000	494.4264	0.3936	0.1609
309.8443	14.81	17200	489.3370	0.3919	0.1610
277.9439	14.98	17400	490.5553	0.3913	0.1610

Framework versions

Transformers 4.38.2
Pytorch 2.2.0+cu118
Datasets 2.16.0
Tokenizers 0.15.2

IshaanHugs/wav2vec2-hiddendrop2

wav2vec2-hiddendrop2

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

IshaanHugs/
wav2vec2-hiddendrop2