Modelpublic

SucharithaS/wav2vec2-large-xls-r-300m-dm32-augmented

sourceHugging Faceapache-2.0updated 5mo agoView on Hugging Face

0likes6downloads

Model Card

wav2vec2-large-xls-r-300m-dm32-augmented

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: 0.6039
—Accuracy: 0.7083
—F1: 0.6901
—Precision: 0.7176
—Recall: 0.7083

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
—trainbatchsize: 4
—evalbatchsize: 4
—seed: 42
—gradientaccumulationsteps: 4
—totaltrainbatch_size: 16
—optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
—lrschedulertype: linear
—lrschedulerwarmup_steps: 200
—num_epochs: 25
—mixedprecisiontraining: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Precision	Recall
No log	2.39	34	0.6977	0.4583	0.4274	0.5139	0.4583
No log	4.77	68	0.6900	0.5833	0.4298	0.3403	0.5833
0.6925	7.16	102	0.6795	0.5833	0.4298	0.3403	0.5833
0.6925	9.54	136	0.6839	0.5833	0.4298	0.3403	0.5833
0.6925	11.93	170	0.6921	0.625	0.625	0.625	0.625
0.6902	14.32	204	0.6799	0.6042	0.4752	0.7642	0.6042
0.6902	16.7	238	0.6752	0.5833	0.4298	0.3403	0.5833
0.6902	19.09	272	0.6352	0.7083	0.7083	0.7083	0.7083
0.6717	21.47	306	0.6492	0.7083	0.6901	0.7176	0.7083
0.6717	23.86	340	0.6039	0.7083	0.6901	0.7176	0.7083

Framework versions

—Transformers 4.37.2
—Pytorch 2.10.0+cu128
—Datasets 4.0.0
—Tokenizers 0.15.2