Modelpublic

alinerodrigues/wav2vec2-large-xlsr-facebook-53-aug-words-phoneme-exp-1-v6

sourceHugging Faceupdated 1y agoView on Hugging Face

0likes10downloads

Model Card

wav2vec2-large-xlsr-facebook-53-aug-words-phoneme-exp-1-v6

This model was trained from scratch on an unknown dataset. It achieves the following results on the evaluation set:

—Loss: 0.4857
—Per Hf Metric: 0.0654
—Per Avg: 0.0896

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: 16
—evalbatchsize: 8
—seed: 42
—gradientaccumulationsteps: 2
—totaltrainbatch_size: 32
—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: 30
—mixedprecisiontraining: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Per Hf Metric	Per Avg
0.383	1.0	102	0.5491	0.0660	0.0891
0.3172	2.0	204	0.5914	0.0671	0.0974
0.2731	3.0	306	0.5024	0.0663	0.0896
0.3224	4.0	408	0.5687	0.0671	0.0946
0.2739	5.0	510	0.4857	0.0662	0.0907
0.2587	6.0	612	0.6277	0.0689	0.0966
0.2521	7.0	714	0.5703	0.0663	0.0923
0.2442	8.0	816	0.6196	0.0633	0.0891
0.2303	9.0	918	0.5667	0.0662	0.0916
0.2066	10.0	1020	0.4967	0.0654	0.0913
0.2365	11.0	1122	0.4858	0.0663	0.0889
0.2041	12.0	1224	0.5368	0.0694	0.0967
0.2105	13.0	1326	0.5410	0.0647	0.0887
0.2016	14.0	1428	0.5634	0.0639	0.0876
0.2133	15.0	1530	0.4990	0.0614	0.0823

Framework versions

—Transformers 4.55.4
—Pytorch 2.8.0+cu126
—Datasets 4.0.0
—Tokenizers 0.21.4