README.md

metadata

license: apache-2.0
base_model: facebook/wav2vec2-xls-r-300m
tags:
  - generated_from_trainer
datasets:
  - fleurs
metrics:
  - wer
model-index:
  - name: wav2vec2-xls-r-300m-mk
    results:
      - task:
          name: Automatic Speech Recognition
          type: automatic-speech-recognition
        dataset:
          name: fleurs
          type: fleurs
          config: mk_mk
          split: test
          args: mk_mk
        metrics:
          - name: Wer
            type: wer
            value: 0.14327357528057136

wav2vec2-xls-r-300m-mk

This model is a fine-tuned version of facebook/wav2vec2-xls-r-300m on Macedonian using the train and validation splits of the FLEURS dataset. It achieves the following results on the evaluation set:

• Loss: 0.1589
• Wer: 0.1433

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: 0.0003
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 16
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• num_epochs: 10
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
3.609	2.33	400	0.3751	0.4184
0.232	4.65	800	0.1694	0.1960
0.0773	6.98	1200	0.1630	0.1598
0.0407	9.3	1600	0.1589	0.1433

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu121
• Datasets 2.16.1
• Tokenizers 0.15.0