Wav2vec2 Xls R 300m Arabic

🚀 XLS-R-300M - Arabic

This is a fine - tuned model for automatic speech recognition in Arabic. It is based on the pre - trained model and achieves certain performance on specific datasets.

🚀 Quick Start

This model is a fine - tuned version of facebook/wav2vec2-xls-r-300m on the MOZILLA - FOUNDATION/COMMON_VOICE_7_0 - AR dataset. It achieves the following results on the evaluation set:

• Loss: 0.4502
• Wer: 0.4783

✨ Features

• Multilingual Adaptability: Based on a large - scale pre - trained model, it can potentially adapt to multiple languages with fine - tuning.
• Performance on Specific Datasets: Achieves certain accuracy on the Arabic dataset of Common Voice 7 and other related speech recognition datasets.

📚 Documentation

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: 7.5e - 05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 4
• 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: 2000
• num_epochs: 5.0
• mixed_precision_training: Native AMP

Training results

Framework versions

• Transformers 4.16.0.dev0
• Pytorch 1.10.1+cu102
• Datasets 1.17.1.dev0
• Tokenizers 0.11.0

📄 License

This model is licensed under the Apache - 2.0 license.

Additional Information

Model Index

• name: XLS - R - 300M - Arabic results:
  • task: name: Automatic Speech Recognition type: automatic - speech - recognition dataset: name: Common Voice 7 type: mozilla - foundation/common_voice_7_0 args: ar metrics:
    • name: Test WER type: wer value: 47.54
    • name: Test CER type: cer value: 17.64
  • task: name: Automatic Speech Recognition type: automatic - speech - recognition dataset: name: Robust Speech Event - Dev Data type: speech - recognition - community - v2/dev_data args: ar metrics:
    • name: Test WER type: wer value: 93.72
  • task: name: Automatic Speech Recognition type: automatic - speech - recognition dataset: name: Robust Speech Event - Test Data type: speech - recognition - community - v2/eval_data args: ar metrics:
    • name: Test WER type: wer value: 92.49

Tags

• ar
• automatic - speech - recognition
• generated_from_trainer
• hf - asr - leaderboard
• mozilla - foundation/common_voice_7_0
• robust - speech - event

Datasets

• mozilla - foundation/common_voice_7_0