You can deploy this model and start earning money today!

KasuleTrevor/
wav2vec2-xls-r-300m-nyn-yogera-v1

Name: wav2vec2-xls-r-300m-nyn-yogera-v1
Availability: InStock

safetensors

wav2vec2

generated_from_trainer

KasuleTrevor/wav2vec2-xls-r-300m-nyn-yogera-v1

finetune

apache-2.0

Views

No views yet

wav2vec2-xls-r-300m-nyn-yogera-v1

This model is a fine-tuned version of KasuleTrevor/wav2vec2-xls-r-300m-nyn-yogera-v1 on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 1.6150
Model Preparation Time: 0.0072
Wer: 0.5460
Cer: 0.1668

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: 5e-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
lr_scheduler_warmup_ratio: 0.033
num_epochs: 100
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Model Preparation Time	Wer	Cer
0.011	0.9965	142	1.5286	0.0072	0.5621	0.1716
0.0098	2.0	285	1.5488	0.0072	0.5573	0.1704
0.0095	2.9965	427	1.5839	0.0072	0.5574	0.1700
0.0077	4.0	570	1.6120	0.0072	0.5549	0.1687
0.0078	4.9965	712	1.6722	0.0072	0.5618	0.1708
0.0077	6.0	855	1.5725	0.0072	0.5545	0.1683
0.0073	6.9965	997	1.6482	0.0072	0.5574	0.1690
0.0067	8.0	1140	1.6416	0.0072	0.5548	0.1691
0.0063	8.9965	1282	1.6240	0.0072	0.5588	0.1708
0.0062	10.0	1425	1.6500	0.0072	0.5561	0.1704
0.006	10.9965	1567	1.6781	0.0072	0.5622	0.1704
0.0054	12.0	1710	1.7039	0.0072	0.5597	0.1697
0.0053	12.9965	1852	1.7232	0.0072	0.5572	0.1696
0.0056	14.0	1995	1.6662	0.0072	0.5549	0.1691
0.0055	14.9965	2137	1.6692	0.0072	0.5555	0.1700
0.0055	16.0	2280	1.5613	0.0072	0.5490	0.1672
0.0057	16.9965	2422	1.6017	0.0072	0.5510	0.1679
0.0051	18.0	2565	1.6355	0.0072	0.5506	0.1679
0.0049	18.9965	2707	1.6400	0.0072	0.5552	0.1687
0.0048	20.0	2850	1.6836	0.0072	0.5498	0.1685
0.0047	20.9965	2992	1.6545	0.0072	0.5531	0.1678
0.0043	22.0	3135	1.7196	0.0072	0.5475	0.1676
0.0045	22.9965	3277	1.6744	0.0072	0.5532	0.1687
0.0046	24.0	3420	1.6859	0.0072	0.5484	0.1688
0.0042	24.9965	3562	1.7156	0.0072	0.5477	0.1677
0.0039	26.0	3705	1.6990	0.0072	0.5509	0.1688
0.0037	26.9965	3847	1.7323	0.0072	0.5504	0.1686
0.0034	28.0	3990	1.7596	0.0072	0.5490	0.1682
0.0038	28.9965	4132	1.7253	0.0072	0.5493	0.1686
0.0041	30.0	4275	1.6648	0.0072	0.5489	0.1673
0.0042	30.9965	4417	1.6150	0.0072	0.5460	0.1668
0.0043	32.0	4560	1.6237	0.0072	0.5487	0.1668
0.0035	32.9965	4702	1.6563	0.0072	0.5534	0.1682
0.0034	34.0	4845	1.7337	0.0072	0.5525	0.1686
0.0035	34.9965	4987	1.7149	0.0072	0.5539	0.1681
0.0035	36.0	5130	1.7267	0.0072	0.5541	0.1680
0.003	36.9965	5272	1.7467	0.0072	0.5521	0.1671
0.0032	38.0	5415	1.7548	0.0072	0.5537	0.1683
0.003	38.9965	5557	1.7468	0.0072	0.5493	0.1671
0.0031	40.0	5700	1.7669	0.0072	0.5514	0.1677
0.0027	40.9965	5842	1.8017	0.0072	0.5499	0.1687
0.0026	42.0	5985	1.7696	0.0072	0.5489	0.1665
0.0028	42.9965	6127	1.7804	0.0072	0.5469	0.1672
0.0027	44.0	6270	1.8494	0.0072	0.5538	0.1687
0.0027	44.9965	6412	1.8231	0.0072	0.5521	0.1681
0.0027	46.0	6555	1.7603	0.0072	0.5505	0.1671

Framework versions

Transformers 4.44.0
Pytorch 2.1.0+cu118
Datasets 2.20.0
Tokenizers 0.19.1