README.md

metadata

license: apache-2.0
tags:
  - generated_from_trainer
datasets:
  - common_voice_16_1
metrics:
  - wer
model-index:
  - name: fine-tune-wav2vec2-large-xls-r-1b-sw
    results:
      - task:
          name: Automatic Speech Recognition
          type: automatic-speech-recognition
        dataset:
          name: common_voice_16_1
          type: common_voice_16_1
          config: ha
          split: train+validation
          args: ha
        metrics:
          - name: Wer
            type: wer
            value: 1

fine-tune-wav2vec2-large-xls-r-1b-sw

This model is a fine-tuned version of facebook/wav2vec2-xls-r-300m on the common_voice_16_1 dataset. It achieves the following results on the evaluation set:

• Loss: 1.8564
• Wer: 1.0

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: 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: 500
• num_epochs: 9

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
No log	2.55	200	2.8195	1.0
4.4719	5.1	400	2.7629	1.0
4.4719	7.64	600	1.8564	1.0

Framework versions

• Transformers 4.27.0
• Pytorch 2.3.0+cu121
• Datasets 2.19.1
• Tokenizers 0.13.3