You can deploy this model and start earning money today!

IshaanHugs/
wav2vec2-hiddendrop3

Name: wav2vec2-hiddendrop3
Availability: InStock

safetensors

wav2vec2

generated_from_trainer

facebook/wav2vec2-xls-r-1b

finetune

apache-2.0

Views

No views yet

wav2vec2-hiddendrop3

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: 505.7838
Wer: 0.3991
Cer: 0.1647

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
1933.963	0.17	200	1881.9603	1.0	1.0
1862.9927	0.34	400	1856.7731	1.0	1.0
1484.8023	0.52	600	1085.7242	0.9964	0.5225
1085.901	0.69	800	785.0924	0.8479	0.3147
893.0736	0.86	1000	746.8273	0.7861	0.2880
911.9345	1.03	1200	626.6774	0.7111	0.2536
768.397	1.21	1400	638.6122	0.6864	0.2507
751.5825	1.38	1600	603.4406	0.6775	0.2421
681.8053	1.55	1800	802.1390	0.7706	0.3118
696.2599	1.72	2000	572.4854	0.6121	0.2224
660.3845	1.89	2200	526.0276	0.6017	0.2169
682.3684	2.07	2400	532.4772	0.5803	0.2114
621.7501	2.24	2600	524.4524	0.5717	0.2093
647.7309	2.41	2800	518.5499	0.5501	0.2038
596.8071	2.58	3000	485.8884	0.5514	0.2041
555.1914	2.76	3200	545.0449	0.5549	0.2061
532.1227	2.93	3400	481.8048	0.5355	0.1988
582.2942	3.1	3600	545.2278	0.5338	0.1986
567.4598	3.27	3800	619.2349	0.5272	0.1979
556.6699	3.44	4000	528.9666	0.5117	0.1945
529.3217	3.62	4200	533.9346	0.5434	0.2020
574.442	3.79	4400	555.8053	0.5206	0.1974
564.9267	3.96	4600	500.5648	0.5140	0.1920
482.9659	4.13	4800	488.3315	0.4990	0.1888
525.8825	4.3	5000	574.4409	0.4963	0.1906
537.3635	4.48	5200	510.6583	0.5055	0.1912
469.9784	4.65	5400	505.7463	0.4881	0.1867
492.8249	4.82	5600	521.2060	0.4951	0.1910
463.9699	4.99	5800	543.5212	0.5057	0.1926
507.6843	5.17	6000	541.4470	0.5056	0.1962
510.1905	5.34	6200	479.5241	0.4810	0.1854
507.6854	5.51	6400	520.8317	0.4973	0.1908
470.14	5.68	6600	541.2523	0.4799	0.1831
500.3853	5.85	6800	549.4141	0.4638	0.1803
465.9087	6.03	7000	512.8759	0.4576	0.1783
478.1954	6.2	7200	559.2225	0.4595	0.1788
494.1672	6.37	7400	613.6085	0.5195	0.2007
448.3457	6.54	7600	465.9614	0.4511	0.1793
448.3134	6.72	7800	453.7521	0.4473	0.1758
459.1452	6.89	8000	513.4709	0.4632	0.1833
462.4243	7.06	8200	460.4988	0.4516	0.1759
440.0421	7.23	8400	535.0063	0.4408	0.1739
447.0938	7.4	8600	495.8009	0.4419	0.1753
437.7381	7.58	8800	474.8371	0.4481	0.1777
446.7597	7.75	9000	474.2127	0.4469	0.1795
405.6966	7.92	9200	537.8873	0.4373	0.1734
453.4284	8.09	9400	459.0724	0.4400	0.1740
443.054	8.27	9600	473.2846	0.4383	0.1730
416.4097	8.44	9800	452.2173	0.4325	0.1735
406.6624	8.61	10000	521.0559	0.4384	0.1754
426.8819	8.78	10200	516.8647	0.4382	0.1736
391.1604	8.95	10400	487.9706	0.4308	0.1712
378.2121	9.13	10600	549.1889	0.4409	0.1752
402.0994	9.3	10800	471.3032	0.4347	0.1743
408.2436	9.47	11000	489.4722	0.4301	0.1710
425.8097	9.64	11200	504.5415	0.4241	0.1700
382.1833	9.81	11400	490.2713	0.4368	0.1749
420.4946	9.99	11600	457.5172	0.4260	0.1735
357.7726	10.16	11800	530.4321	0.4378	0.1729
359.53	10.33	12000	539.8021	0.4332	0.1735
405.7934	10.5	12200	495.9863	0.4220	0.1706
359.8145	10.68	12400	495.0679	0.4137	0.1671
396.3897	10.85	12600	492.0583	0.4089	0.1666
384.8902	11.02	12800	528.7145	0.4197	0.1686
363.3708	11.19	13000	506.8914	0.4298	0.1705
387.44	11.36	13200	454.7159	0.4119	0.1672
343.4508	11.54	13400	508.2368	0.4169	0.1684
431.0554	11.71	13600	463.0938	0.4046	0.1651
379.5059	11.88	13800	465.7052	0.4023	0.1653
381.347	12.05	14000	489.7500	0.4109	0.1670
366.2464	12.23	14200	475.6771	0.4124	0.1680
365.56	12.4	14400	494.1970	0.4093	0.1671
366.7099	12.57	14600	485.0564	0.4177	0.1704
339.8638	12.74	14800	495.2524	0.4109	0.1672
365.75	12.91	15000	487.8662	0.4020	0.1648
289.899	13.09	15200	496.7301	0.4012	0.1645
358.4975	13.26	15400	475.8522	0.4056	0.1660
349.8539	13.43	15600	471.1264	0.4041	0.1666
399.4385	13.6	15800	498.9843	0.4041	0.1654
344.7482	13.78	16000	478.8123	0.3999	0.1648
362.1367	13.95	16200	498.5487	0.3995	0.1650
318.3352	14.12	16400	523.2060	0.3996	0.1651
372.5933	14.29	16600	508.3610	0.4034	0.1656
311.8611	14.46	16800	512.0081	0.4016	0.1649
351.7635	14.64	17000	508.1662	0.4014	0.1648
335.4864	14.81	17200	506.4462	0.3996	0.1647
304.3739	14.98	17400	505.7838	0.3991	0.1647

Framework versions

Transformers 4.38.2
Pytorch 2.2.0+cu118
Datasets 2.16.0
Tokenizers 0.15.2

IshaanHugs/wav2vec2-hiddendrop3

wav2vec2-hiddendrop3

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

IshaanHugs/
wav2vec2-hiddendrop3