metadata

language:
  - ku
license: apache-2.0
base_model: openai/whisper-small
tags:
  - generated_from_trainer
datasets:
  - mozilla-foundation/common_voice_11_0
metrics:
  - wer
model-index:
  - name: Whisper Small Kur - Rizgan Gerdenzeri
    results:
      - task:
          name: Automatic Speech Recognition
          type: automatic-speech-recognition
        dataset:
          name: Common Voice 17.0
          type: mozilla-foundation/common_voice_11_0
          config: kmr
          split: None
          args: 'config: kmr, split: test'
        metrics:
          - name: Wer
            type: wer
            value: 35.26864147088866

Whisper Small Kur - Rizgan Gerdenzeri

This model is a fine-tuned version of openai/whisper-small on the Common Voice 17.0 dataset. It achieves the following results on the evaluation set:

Loss: 0.5986
Wer: 35.2686

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: 4000
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.3355	1.7699	1000	0.4746	40.3146
0.0921	3.5398	2000	0.4746	36.7845
0.0142	5.3097	3000	0.5598	36.6251
0.004	7.0796	4000	0.5986	35.2686

Framework versions

Transformers 4.43.3
Pytorch 2.4.0+cu121
Datasets 2.20.0
Tokenizers 0.19.1