metadata

license: apache-2.0
tags:
  - whisper-event
  - generated_from_trainer
datasets:
  - NbAiLab/NCC_S
metrics:
  - wer
model-index:
  - name: Whisper Base Norwegian
    results:
      - task:
          name: Automatic Speech Recognition
          type: automatic-speech-recognition
        dataset:
          name: NbAiLab/NCC_S
          type: NbAiLab/NCC_S
          config: 'no'
          split: validation
          args: 'no'
        metrics:
          - name: Wer
            type: wer
            value: 16.017052375152254

Whisper Base Norwegian

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

Loss: 0.3290
Wer: 16.0171

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: 64
eval_batch_size: 32
seed: 42
gradient_accumulation_steps: 2
total_train_batch_size: 128
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: constant_with_warmup
lr_scheduler_warmup_steps: 500
training_steps: 5000
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.8528	0.2	1000	0.4402	22.5639
0.7558	0.4	2000	0.3861	19.7016
0.721	0.6	3000	0.3545	17.8441
0.6803	0.8	4000	0.3364	16.5652
0.6557	1.0	5000	0.3290	16.0171

Framework versions

Transformers 4.26.0.dev0
Pytorch 1.13.0+cu117
Datasets 2.7.1.dev0
Tokenizers 0.13.2