README.md

metadata

language:
  - ga-IE
license: apache-2.0
tags:
  - automatic-speech-recognition
  - mozilla-foundation/common_voice_7_0
  - generated_from_trainer
  - ga-IE
  - robust-speech-event
  - model_for_talk
  - hf-asr-leaderboard
datasets:
  - mozilla-foundation/common_voice_7_0
model-index:
  - name: XLS-R-300M - Irish
    results:
      - task:
          name: Automatic Speech Recognition
          type: automatic-speech-recognition
        dataset:
          name: Common Voice 7
          type: mozilla-foundation/common_voice_7_0
          args: ga-IE
        metrics:
          - name: Test WER
            type: wer
            value: 103.54
          - name: Test CER
            type: cer
            value: 326.923

wav2vec2-large-xls-r-300m-irish

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

• Loss: 1.1647
• Wer: 0.7296

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: 7e-05
• train_batch_size: 64
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 256
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 200
• num_epochs: 300.0
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
2.9022	124.94	500	2.7763	0.9824
1.5112	249.94	1000	1.1736	0.7405

Framework versions

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