wav2vec2-large-xls-r-300m-kor-lr-3e-3

This model is a fine-tuned version of teddy322/wav2vec2-large-xls-r-300m-kor-lr-3e-3 on the zeroth_korean_asr dataset. It achieves the following results on the evaluation set:

• eval_loss: 1.7324
• eval_wer: 0.9395
• eval_runtime: 128.5875
• eval_samples_per_second: 3.554
• eval_steps_per_second: 0.451
• epoch: 12.36
• step: 4400

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: 0.003
• 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: 20
• mixed_precision_training: Native AMP

Framework versions

• Transformers 4.11.3
• Pytorch 1.10.0+cu113
• Datasets 1.18.3
• Tokenizers 0.10.3