README.md

metadata

language:
  - nl
license: apache-2.0
tags:
  - hf-asr-leaderboard
  - generated_from_trainer
datasets:
  - mozilla-foundation/common_voice_11_0
metrics:
  - wer
model-index:
  - name: Whisper Dutch - RTL
    results:
      - task:
          name: Automatic Speech Recognition
          type: automatic-speech-recognition
        dataset:
          name: Common Voice 11.0
          type: mozilla-foundation/common_voice_11_0
          config: dutch
          split: None
          args: 'config: nl, split: test'
        metrics:
          - name: Wer
            type: wer
            value: 37.50805849185964

Whisper Dutch - RTL

This model is a fine-tuned version of openai/whisper-small on the Common Voice 11.0 dataset. It achieves the following results on the evaluation set:

• Loss: 0.1790
• Wer: 37.5081

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: 1e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• gradient_accumulation_steps: 2
• 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
• training_steps: 4000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.1238	0.78	1000	0.2017	19.8254
0.0548	1.56	2000	0.1829	35.4625
0.0259	2.34	3000	0.1795	43.1853
0.0131	3.12	4000	0.1790	37.5081

Framework versions

• Transformers 4.28.1
• Pytorch 2.0.0+cu117
• Datasets 2.12.0
• Tokenizers 0.13.3