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-tiny
model-index:
- name: Whisper Tiny 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: 65.19764801769809
name: Wer
Whisper Tiny Tr - Canberk Kandemir
This model is a fine-tuned version of openai/whisper-tiny on the Common Voice 11.0 dataset. It achieves the following results on the evaluation set:
- Loss: 0.5050
- Wer: 65.1976
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: 3e-05
- train_batch_size: 16
- eval_batch_size: 16
- 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: cosine_with_restarts
- lr_scheduler_warmup_steps: 500
- training_steps: 4000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.373 | 0.89 | 1000 | 0.5859 | 60.4296 |
| 0.2174 | 1.77 | 2000 | 0.5236 | 49.7332 |
| 0.1194 | 2.66 | 3000 | 0.5080 | 61.9743 |
| 0.0832 | 3.54 | 4000 | 0.5050 | 65.1976 |
Framework versions
- Transformers 4.37.2
- Pytorch 2.1.0+cu121
- Datasets 2.17.0
- Tokenizers 0.15.2