Wav2vec2 Xlsr Lithuanian

🚀 sammy786/wav2vec2-xlsr-lithuanian

This model is a fine - tuned version of facebook/wav2vec2-xls-r-1b on the MOZILLA-FOUNDATION/COMMON_VOICE_8_0 - lt dataset. It can be used for automatic speech recognition tasks, providing a relatively accurate speech - to - text conversion solution.

✨ Features

• Fine - tuned on the MOZILLA - FOUNDATION/COMMON_VOICE_8_0 - lt dataset, improving performance on Lithuanian speech recognition.
• Achieved specific loss and WER results on the evaluation set, demonstrating its effectiveness.

📦 Installation

No installation steps are provided in the original document.

💻 Usage Examples

No code examples are provided in the original document.

📚 Documentation

Model description

"facebook/wav2vec2-xls-r-1b" was finetuned.

Intended uses & limitations

More information needed

Training and evaluation data

Training data - Common voice Finnish train.tsv, dev.tsv and other.tsv

Training procedure

For creating the train dataset, all possible datasets were appended and 90 - 10 split was used.

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 0.000045637994662983496
• train_batch_size: 8
• eval_batch_size: 16
• seed: 13
• gradient_accumulation_steps: 4
• total_train_batch_size: 32
• optimizer: Adam with betas=(0.9,0.999) and epsilon = 1e - 08
• lr_scheduler_type: cosine_with_restarts
• lr_scheduler_warmup_steps: 500
• num_epochs: 40
• mixed_precision_training: Native AMP

Training results

Framework versions

• Transformers 4.16.0.dev0
• Pytorch 1.10.0+cu102
• Datasets 1.17.1.dev0
• Tokenizers 0.10.3

Evaluation Commands

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

python eval.py --model_id sammy786/wav2vec2-xlsr-lithuanian --dataset mozilla-foundation/common_voice_8_0 --config lt --split test

🔧 Technical Details

Model Performance

It achieves the following results on evaluation set (which is 10 percent of train data set merged with other and dev datasets):

• Loss: 13.1811
• Wer: 24.2570

📄 License

This model is licensed under the Apache - 2.0 license.