metadata
language:
- tr
license: apache-2.0
tags:
- hf-asr-leaderboard
- generated_from_trainer
datasets:
- mozilla-foundation/common_voice_11_0
metrics:
- wer
base_model: openai/whisper-small
model-index:
- name: Whisper Small Tr - Canberk Kandemir
results:
- task:
type: automatic-speech-recognition
name: Automatic Speech Recognition
dataset:
name: Common Voice 11.0
type: mozilla-foundation/common_voice_11_0
config: tr
split: None
args: 'config: tr, split: test'
metrics:
- type: wer
value: 43.06339873086104
name: Wer
Whisper Small Tr
This model is a fine-tuned version of openai/whisper-small on the Common Voice 11.0 dataset. It achieves the following results on the evaluation set:
- Loss: 0.5432
- Wer: 43.0634
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: 7e-05
- train_batch_size: 16
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 1000
- training_steps: 5000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.4399 | 0.44 | 500 | 0.6307 | 61.0351 |
| 0.4322 | 0.89 | 1000 | 0.6820 | 58.6909 |
| 0.2857 | 1.33 | 1500 | 0.6496 | 54.3867 |
| 0.2839 | 1.77 | 2000 | 0.6088 | 49.6497 |
| 0.1467 | 2.21 | 2500 | 0.5813 | 47.3346 |
| 0.1268 | 2.66 | 3000 | 0.5647 | 46.1315 |
| 0.0711 | 3.1 | 3500 | 0.5532 | 44.8196 |
| 0.0658 | 3.54 | 4000 | 0.5444 | 43.4670 |
| 0.0601 | 3.99 | 4500 | 0.5372 | 43.4146 |
| 0.0304 | 4.43 | 5000 | 0.5432 | 43.0634 |
Framework versions
- Transformers 4.37.2
- Pytorch 2.1.0+cu121
- Datasets 2.17.0
- Tokenizers 0.15.2