You can deploy this model and start earning money today!

teddy322/
wav2vec2-large-xls-r-300m-korean

Name: wav2vec2-large-xls-r-300m-korean
Availability: InStock

automatic-speech-recognition

generated_from_trainer

zeroth_korean_asr

apache-2.0

endpoints_compatible

Views

No views yet

wav2vec2-large-xls-r-300m-korean

This model is a fine-tuned version of teddy322/wav2vec2-large-xls-r-300m-korean on the zeroth_korean_asr dataset. It achieves the following results on the evaluation set:

Loss: 0.4474
Wer: 0.3320

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: 16
eval_batch_size: 8
seed: 42
gradient_accumulation_steps: 2
total_train_batch_size: 32
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 500
num_epochs: 10
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.1683	1.12	400	0.4871	0.4144
0.2177	2.25	800	0.5225	0.4552
0.1939	3.37	1200	0.5300	0.4456
0.1432	4.49	1600	0.4704	0.3850
0.1047	5.62	2000	0.4951	0.3960
0.0864	6.74	2400	0.4617	0.3638
0.0686	7.87	2800	0.4477	0.3393
0.0538	8.99	3200	0.4474	0.3320

Framework versions

Transformers 4.11.3
Pytorch 1.10.0+cu113
Datasets 1.18.3
Tokenizers 0.10.3