Wav2vec2 Large Xls R 300m Hsb V1

🚀 wav2vec2-large-xls-r-300m-hsb-v1

This model is a fine - tuned version of facebook/wav2vec2-xls-r-300m on the MOZILLA - FOUNDATION/COMMON_VOICE_8_0 - HSB dataset. It is designed for automatic speech recognition in the Upper Sorbian (HSB) language, offering high - quality performance for related tasks.

✨ Features

• Language Support: Specialized for the Upper Sorbian (HSB) language in automatic speech recognition.
• Fine - Tuned Model: Based on the pre - trained facebook/wav2vec2-xls-r-300m model, fine - tuned on the MOZILLA - FOUNDATION/COMMON_VOICE_8_0 - HSB dataset.
• Comprehensive Evaluation: Evaluated on multiple datasets with various metrics such as WER and CER.

📦 Installation

The document doesn't provide specific installation steps, so this section is skipped.

💻 Usage Examples

Basic Usage

The document doesn't provide basic usage code examples, so this part is skipped.

Advanced Usage

The document doesn't provide advanced usage code examples, so this part is skipped.

📚 Documentation

Evaluation Commands

1. To evaluate on mozilla - foundation/common_voice_8_0 with test split

python eval.py --model_id DrishtiSharma/wav2vec2-large-xls-r-300m-hsb-v1 --dataset mozilla-foundation/common_voice_8_0 --config hsb --split test --log_outputs

2. To evaluate on speech - recognition - community - v2/dev_data Upper Sorbian language isn't available in speech - recognition - community - v2/dev_data

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 0.00045
• 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: 50
• mixed_precision_training: Native AMP

Training results

Framework versions

• Transformers 4.16.1
• Pytorch 1.10.0+cu111
• Datasets 1.18.2
• Tokenizers 0.11.0

🔧 Technical Details

The model is a fine - tuned version of facebook/wav2vec2-xls-r-300m on the MOZILLA - FOUNDATION/COMMON_VOICE_8_0 - HSB dataset. It achieves certain results on the evaluation set, with a loss of 0.5684 and a Wer of 0.4402. During training, specific hyperparameters were used, and the training results are presented in a table, showing the changes in training loss, validation loss, and Wer over epochs and steps.

📄 License

The model is licensed under the Apache - 2.0 license.

Additional Information