Modelpublic

tadgeis/xls-r-300m-ckt-staged-bible-radio-to-chuklang-source-eval-stage2

sourceHugging Faceapache-2.0updated 4mo agoView on Hugging Face

0likes12downloads

Model Card

xls-r-300m-ckt-staged-bible-radio-to-chuklang-source-eval-stage2

This model is a fine-tuned version of facebook/wav2vec2-xls-r-300m.

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: 1e-05
—trainbatchsize: 4
—evalbatchsize: 4
—seed: 42
—optimizer: Use OptimizerNames.ADAMWTORCHFUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
—lrschedulertype: linear
—num_epochs: 12
—mixedprecisiontraining: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer	Cer
0.2439	0.3030	50	1.1534	0.8667	0.2442
0.2182	0.6061	100	1.1732	0.8635	0.2405
0.2084	0.9091	150	1.1908	0.8635	0.2427
0.1919	1.2121	200	1.2112	0.8667	0.2407
0.1958	1.5152	250	1.2202	0.8667	0.2418
0.1887	1.8182	300	1.2307	0.8667	0.2405
0.165	2.1212	350	1.2353	0.8635	0.2405

Framework versions

—Transformers 4.57.1
—Pytorch 2.11.0+cu130
—Datasets 4.8.5
—Tokenizers 0.22.2