Modelpublic

KasuleTrevor/Runynkore_speech_to_intent_multilingual_xlsr

sourceHugging Faceupdated 1y agoView on Hugging Face

0likes3downloads

Model Card

Runynkorespeechtointentmultilingual_xlsr

This model is a fine-tuned version of KasuleTrevor/wav2vec2-xls-r-300m-multilingual_filtered-yogera-v3 on an unknown dataset. It achieves the following results on the evaluation set:

—Loss: 0.1395
—Accuracy: 0.9722
—Precision: 0.9746
—Recall: 0.9722
—F1: 0.9717

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: 0.0001
—trainbatchsize: 32
—evalbatchsize: 8
—seed: 42
—gradientaccumulationsteps: 2
—totaltrainbatch_size: 64
—optimizer: Use adamwtorch with betas=(0.9,0.999) and epsilon=1e-08 and optimizerargs=No additional optimizer arguments
—lrschedulertype: cosine
—lrschedulerwarmup_steps: 500
—num_epochs: 20
—mixedprecisiontraining: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1
No log	1.0	88	2.8125	0.1405	0.1094	0.1405	0.0721
2.9239	2.0	176	1.2254	0.7149	0.6086	0.7149	0.6258
1.8873	3.0	264	0.1759	0.9796	0.9810	0.9796	0.9798
0.3484	4.0	352	0.1055	0.9796	0.9807	0.9796	0.9797
0.1375	5.0	440	0.1020	0.9796	0.9807	0.9796	0.9797
0.1062	6.0	528	0.0892	0.9817	0.9828	0.9817	0.9817
0.0891	7.0	616	0.0808	0.9837	0.9843	0.9837	0.9837
0.0687	8.0	704	0.0671	0.9817	0.9830	0.9817	0.9816
0.0687	9.0	792	0.0760	0.9796	0.9802	0.9796	0.9796
0.0501	10.0	880	0.0818	0.9817	0.9826	0.9817	0.9817
0.0411	11.0	968	0.0820	0.9857	0.9865	0.9857	0.9858
0.0259	12.0	1056	0.0812	0.9796	0.9800	0.9796	0.9796
0.032	13.0	1144	0.0794	0.9817	0.9822	0.9817	0.9817
0.0266	14.0	1232	0.0976	0.9817	0.9826	0.9817	0.9817
0.0211	15.0	1320	0.1037	0.9796	0.9806	0.9796	0.9797
0.0178	16.0	1408	0.0949	0.9796	0.9802	0.9796	0.9797
0.0178	17.0	1496	0.0901	0.9817	0.9827	0.9817	0.9817
0.0164	18.0	1584	0.0936	0.9817	0.9825	0.9817	0.9817
0.017	19.0	1672	0.0939	0.9817	0.9827	0.9817	0.9817
0.0147	20.0	1760	0.0940	0.9817	0.9827	0.9817	0.9817

Framework versions

—Transformers 4.51.3
—Pytorch 2.1.0+cu118
—Datasets 3.6.0
—Tokenizers 0.21.1