metadata

language:
  - el
tags:
  - >-
    hf-asr-leaderboard, whisper-medium, mozilla-foundation/common_voice_11_0,
    greek, whisper-event
  - generated_from_trainer
datasets:
  - google/fleurs
metrics:
  - wer
model-index:
  - name: Whisper Medium El - Greek One
    results:
      - task:
          name: Automatic Speech Recognition
          type: automatic-speech-recognition
        dataset:
          name: Google FLEURS
          type: google/fleurs
          config: el_gr
          split: test
          args: el_gr
        metrics:
          - name: Wer
            type: wer
            value: 15.584586962259174

Whisper Medium El - Greek One

This model is a fine-tuned version of openai/medium-medium on the Google FLEURS dataset. It achieves the following results on the evaluation set:

Loss: 0.2864
Wer: 15.5846

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: 20
eval_batch_size: 8
seed: 42
gradient_accumulation_steps: 2
total_train_batch_size: 40
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 500
training_steps: 2000
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.0006	12.02	1000	0.2718	15.4394
0.0003	24.04	2000	0.2864	15.5846

Framework versions

Transformers 4.26.0.dev0
Pytorch 1.14.0.dev20221206+cu116
Datasets 2.7.1.dev0
Tokenizers 0.13.2