This model is a fine-tuned version of facebook/mms-1b-all on the jssaluja/rajinder_singh dataset. It achieves the following results on the evaluation set:

Loss: 0.3182
Wer: 0.3554
Wil: 0.5543
Mer: 0.3478
Cer: 0.0925
Ncer: 0.0473
Ccer: 0.0979

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: 0.001
train_batch_size: 8
eval_batch_size: 8
seed: 42
optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
num_epochs: 10
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Ccer	Cer	Validation Loss	Mer	Ncer	Wer	Wil
0.4255	1.0	7548	0.1226	0.1166	0.3802	0.4426	0.0577	0.4511	0.6702
0.2044	2.0	15096	0.1070	0.1045	0.3494	0.4009	0.0517	0.4139	0.6196
0.1835	3.0	22644	0.1083	0.1051	0.3483	0.3999	0.0517	0.4115	0.6186
0.1683	4.0	30192	0.1093	0.1047	0.3476	0.3954	0.0502	0.4022	0.6130
0.1541	5.0	37740	0.1052	0.0987	0.3314	0.3707	0.0500	0.3792	0.5842
0.1439	6.0	45288	0.1012	0.0962	0.3275	0.3624	0.0498	0.3699	0.5721
0.1349	7.0	52836	0.1014	0.0964	0.3306	0.3664	0.0476	0.3728	0.5776
0.1288	8.0	60384	0.0986	0.0930	0.3096	0.3510	0.0471	0.3584	0.5593
0.1208	9.0	67932	0.1009	0.0956	0.3253	0.3604	0.0494	0.3674	0.5696
0.1149	10.0	75480	0.3182	0.3554	0.5543	0.3478	0.0925	0.0473	0.0979

Framework versions

Transformers 4.52.4
Pytorch 2.6.0+cu124
Datasets 3.6.0
Tokenizers 0.21.2

SidakAI/fb-mms-1b-ccer-jssaluja_rajinder_singh-epochs-10-test-datasets-10-20251026_094435

jssaluja/fb-mms-1b-ccer-jssaluja_rajinder_singh-epochs-10-test-datasets-10-20251026_094435

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

SidakAI/
fb-mms-1b-ccer-jssaluja_rajinder_singh-epochs-10-test-datasets-10-20251026_094435