metadata

base_model: openai/whisper-base
datasets:
  - mozilla-foundation/common_voice_11_0
language:
  - id
library_name: transformers
license: apache-2.0
metrics:
  - wer
tags:
  - generated_from_trainer
model-index:
  - name: whisper-base-id
    results:
      - task:
          type: automatic-speech-recognition
          name: Automatic Speech Recognition
        dataset:
          name: Common Voice 11.0
          type: mozilla-foundation/common_voice_11_0
          config: id
          split: test
          args: 'config: id, split: test'
        metrics:
          - type: wer
            value: 28.978092004279272
            name: Wer

whisper-base-id

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

Loss: 0.4661
Wer: 28.9781

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: 8
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 500
training_steps: 5000
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.3685	1.9305	1000	0.3951	28.4153
0.1421	3.8610	2000	0.3944	28.3269
0.0494	5.7915	3000	0.4211	28.4153
0.0176	7.7220	4000	0.4514	30.2712
0.0105	9.6525	5000	0.4661	28.9781

Framework versions

Transformers 4.44.2
Pytorch 2.4.0+cu121
Datasets 3.0.0
Tokenizers 0.19.1