wav2vec2-large-xlsr-53-german-finetuned-ks-de

This model is a fine-tuned version of jonatasgrosman/wav2vec2-large-xlsr-53-german on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.8681
• Accuracy: 0.6667

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: 3e-05
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 128
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	1	1.9490	0.0833
No log	2.0	2	1.9128	0.25
No log	3.0	3	1.8861	0.5833
No log	4.0	4	1.8681	0.6667
No log	5.0	5	1.8590	0.6667

Framework versions

• Transformers 4.20.1
• Pytorch 1.11.0+cu113
• Datasets 2.3.2
• Tokenizers 0.12.1