wav2vec2-large-mms-1b-korean-colab

This model is a fine-tuned version of weekcircle/wav2vec2-large-mms-1b-korean-colab on the common_voice_13_0 dataset. It achieves the following results on the evaluation set:

• Loss: 8.8258
• Wer: 0.9960

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: 0.001
• 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: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
2.4313	2.63	100	7.9123	0.9839
2.3616	5.26	200	7.9118	0.9930
1.859	7.89	300	7.9977	0.9909
1.4135	10.53	400	8.3395	1.0040
1.1407	13.16	500	8.5900	0.9940
0.9639	15.79	600	8.6300	0.9950
0.7991	18.42	700	8.8258	0.9960

Framework versions

• Transformers 4.31.0.dev0
• Pytorch 2.0.1+cu118
• Datasets 2.13.1
• Tokenizers 0.13.3