wav2vec2-large-xls-r-300m-zhhk

This model is a fine-tuned version of ctl/wav2vec2-large-xlsr-cantonese on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.5094
• Cer: 0.3540

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
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 16
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 100
• num_epochs: 1
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Cer
0.6545	0.32	400	1.6247	0.3646
0.621	0.64	800	1.5691	0.3588
0.8621	0.96	1200	1.5094	0.3540

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu121
• Datasets 2.15.0
• Tokenizers 0.15.0