German Pretrained

🚀 german_pretrained

This model is a fine - tuned version of flozi00/wav2vec-xlsr-german on the None dataset. It offers evaluation results including a Loss of 3.9812 and a Wer of 1.0, which shows its performance in relevant tasks.

🚀 Quick Start

This section provides a quick overview of the model. The german_pretrained model is fine - tuned from flozi00/wav2vec-xlsr-german and has specific performance metrics on the evaluation set.

📚 Documentation

Model description

This model is a fine - tuned version of flozi00/wav2vec-xlsr-german on the None dataset. It achieves the following results on the evaluation set:

• Loss: 3.9812
• Wer: 1.0

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: 5
• num_epochs: 30

Training results

Framework versions

• Transformers 4.11.3
• Pytorch 1.10.0+cu102
• Datasets 1.13.3
• Tokenizers 0.10.3

📄 License

This model is licensed under the Apache 2.0 license.