You can deploy this model and start earning money today!

alinerodrigues/
wav2vec2-large-xlsr-facebook-300m-words-phoneme-exp-1

Name: wav2vec2-large-xlsr-facebook-300m-words-phoneme-exp-1
Availability: InStock

pytorch

wav2vec2

generated_from_trainer

apache-2.0

Views

No views yet

wav2vec2-large-xlsr-facebook-300m-words-phoneme-exp-1

This model is a fine-tuned version of facebook/wav2vec2-xls-r-300m on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 0.4032
Per: 0.1745

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: 3e-05
train_batch_size: 16
eval_batch_size: 8
seed: 42
gradient_accumulation_steps: 2
total_train_batch_size: 32
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 200
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Per
9.2635	1.0	101	4.4060	1.0
3.8857	2.0	203	3.2531	1.0
3.1446	3.0	304	3.0293	1.0
2.9876	4.0	406	2.8649	1.0
2.8076	5.0	507	2.6248	1.0
2.6495	6.0	609	2.5289	1.0
2.5239	7.0	710	2.4092	0.9996
2.4091	8.0	812	2.2508	0.8973
2.239	9.0	913	2.0762	0.8385
2.0924	10.0	1015	1.8913	0.7653
1.9069	11.0	1116	1.6510	0.6886
1.7084	12.0	1218	1.4183	0.6512
1.5347	13.0	1319	1.2680	0.6420
1.3838	14.0	1421	1.1378	0.5671
1.2594	15.0	1522	0.9868	0.5421
1.1283	16.0	1624	0.9106	0.5161
1.0664	17.0	1725	0.8059	0.4550
0.9531	18.0	1827	0.7782	0.4529
0.8917	19.0	1928	0.7027	0.3999
0.828	20.0	2030	0.6596	0.3670
0.7783	21.0	2131	0.6449	0.3115
0.7226	22.0	2233	0.5703	0.3269
0.6976	23.0	2334	0.6207	0.2894
0.6572	24.0	2436	0.5328	0.2828
0.6208	25.0	2537	0.5301	0.2655
0.5892	26.0	2639	0.5276	0.2536
0.5875	27.0	2740	0.5158	0.2458
0.5567	28.0	2842	0.4809	0.2711
0.5251	29.0	2943	0.4701	0.2618
0.4966	30.0	3045	0.4725	0.2488
0.4727	31.0	3146	0.4752	0.2695
0.4771	32.0	3248	0.4514	0.2422
0.4453	33.0	3349	0.4577	0.2285
0.4288	34.0	3451	0.4806	0.2130
0.4182	35.0	3552	0.4253	0.2329
0.3927	36.0	3654	0.4855	0.2231
0.3896	37.0	3755	0.4313	0.2144
0.4072	38.0	3857	0.4433	0.2210
0.3732	39.0	3958	0.4225	0.2301
0.397	40.0	4060	0.4413	0.2215
0.3586	41.0	4161	0.4333	0.2231
0.3744	42.0	4263	0.4838	0.2310
0.3596	43.0	4364	0.4447	0.2208
0.3364	44.0	4466	0.4424	0.2144
0.3507	45.0	4567	0.4375	0.2140
0.3282	46.0	4669	0.4237	0.2007
0.3122	47.0	4770	0.4686	0.1968
0.3273	48.0	4872	0.4560	0.2048
0.2831	49.0	4973	0.4552	0.2057
0.3027	50.0	5075	0.4762	0.1991
0.2696	51.0	5176	0.4266	0.2017
0.2748	52.0	5278	0.4120	0.1980
0.2704	53.0	5379	0.4554	0.2028
0.2758	54.0	5481	0.4817	0.1969
0.2525	55.0	5582	0.4032	0.1745
0.262	56.0	5684	0.4515	0.1836
0.2448	57.0	5785	0.5141	0.1837
0.2588	58.0	5887	0.4413	0.1984
0.2423	59.0	5988	0.5094	0.1789
0.2334	60.0	6090	0.4503	0.1686
0.2378	61.0	6191	0.5023	0.1772
0.2464	62.0	6293	0.4704	0.1625
0.2213	63.0	6394	0.4438	0.1802
0.2099	64.0	6496	0.4301	0.2034
0.2414	65.0	6597	0.4739	0.1902
0.2068	66.0	6699	0.4536	0.2014
0.2128	67.0	6800	0.4507	0.2135
0.1925	68.0	6902	0.4873	0.2014
0.1916	69.0	7003	0.4212	0.2060
0.1857	70.0	7105	0.4306	0.1841
0.1789	71.0	7206	0.4401	0.2003
0.2011	72.0	7308	0.4712	0.2078
0.1922	73.0	7409	0.4643	0.2221
0.1888	74.0	7511	0.4818	0.2112
0.1902	75.0	7612	0.4573	0.2190

Framework versions

Transformers 4.28.0
Pytorch 2.6.0+cu124
Datasets 3.6.0
Tokenizers 0.13.3

alinerodrigues/wav2vec2-large-xlsr-facebook-300m-words-phoneme-exp-1

wav2vec2-large-xlsr-facebook-300m-words-phoneme-exp-1

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

alinerodrigues/
wav2vec2-large-xlsr-facebook-300m-words-phoneme-exp-1