metadata

library_name: transformers
language:
  - en
license: apache-2.0
base_model: openai/whisper-base
tags:
  - generated_from_trainer
datasets:
  - DTU54DL/common-accent
metrics:
  - wer
  - precision
  - recall
  - f1
model-index:
  - name: Whisper Base CA
    results:
      - task:
          name: Automatic Speech Recognition
          type: automatic-speech-recognition
        dataset:
          name: Common Accent
          type: DTU54DL/common-accent
        metrics:
          - name: Wer
            type: wer
            value: 0.26376410965215386
          - name: Precision
            type: precision
            value: 0.8083025813102722
          - name: Recall
            type: recall
            value: 0.8232867121696472
          - name: F1
            type: f1
            value: 0.8149744272232056

Whisper Base CA

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

Loss: 0.7230
Wer Ortho: 30.5998
Wer: 0.2638
Cer: 0.1320
Precision: 0.8083
Recall: 0.8233
F1: 0.8150

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
optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: constant_with_warmup
lr_scheduler_warmup_steps: 50
training_steps: 500
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer Ortho	Wer	Cer	Precision	Recall	F1
0.1367	0.8	500	0.7230	30.5998	0.2638	0.1320	0.8083	0.8233	0.8150

Framework versions

Transformers 4.47.1
Pytorch 2.5.1+cu121
Datasets 3.2.0
Tokenizers 0.21.0