wav2vec2-large-xls-r-300m-korean_speed – AI Model by Gummybear05 | AlphaNeural AI

You can deploy this model and start earning money today!

Gummybear05/
wav2vec2-large-xls-r-300m-korean_speed

like

0

automatic-speech-recognition

generated_from_trainer

facebook/wav2vec2-xls-r-300m

endpoints_compatible

Views

No views yet

wav2vec2-large-xls-r-300m-korean_speed

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: 4.3672
Cer: 98.3846

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: 0.0003
train_batch_size: 8
eval_batch_size: 8
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 100
num_epochs: 3
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Cer
34.1959	0.13	200	5.0888	100.0
5.0127	0.26	400	4.8530	100.0
4.8574	0.39	600	4.8649	100.0
4.8596	0.52	800	4.6823	100.0
4.7742	0.64	1000	4.9200	100.0
4.7557	0.77	1200	4.6647	100.0
4.7367	0.9	1400	4.6535	100.0
4.7075	1.03	1600	4.6910	100.0
4.6994	1.16	1800	4.7617	100.0
4.6886	1.29	2000	4.7125	100.0
4.6495	1.42	2200	4.7181	100.0
4.651	1.55	2400	4.7701	100.0
4.5945	1.68	2600	4.7640	100.0
4.5589	1.81	2800	5.2212	100.0
4.5125	1.93	3000	4.6738	99.9178
4.4526	2.06	3200	4.6993	99.8649
4.3971	2.19	3400	4.6336	99.7063
4.3061	2.32	3600	4.6569	99.2775
4.2643	2.45	3800	4.5247	99.4243
4.1763	2.58	4000	4.5722	99.1189
4.106	2.71	4200	4.4703	99.0778
4.0419	2.84	4400	4.4161	97.9617
4.0283	2.97	4600	4.3672	98.3846

Framework versions

Transformers 4.35.2
Pytorch 2.1.0+cu118
Datasets 2.15.0
Tokenizers 0.15.0