metadata
language:
- uk
license: apache-2.0
base_model: openai/whisper-small
tags:
- whisper-event
- generated_from_trainer
datasets:
- mozilla-foundation/common_voice_16_0
metrics:
- wer
model-index:
- name: Whisper Small Ukrainian
results:
- task:
name: Automatic Speech Recognition
type: automatic-speech-recognition
dataset:
name: mozilla-foundation/common_voice_16_0 uk
type: mozilla-foundation/common_voice_16_0
config: uk
split: test
args: uk
metrics:
- name: Wer
type: wer
value: 22.412355692464214
Whisper Small Ukrainian
This model is a fine-tuned version of openai/whisper-small on the mozilla-foundation/common_voice_16_0 uk dataset. It achieves the following results on the evaluation set:
- Loss: 0.3487
- Wer: 22.4124
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-06
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 50
- training_steps: 2500
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.3468 | 1.08 | 500 | 0.3877 | 24.7185 |
| 0.2915 | 3.03 | 1000 | 0.3639 | 23.2277 |
| 0.2671 | 4.11 | 1500 | 0.3570 | 22.7472 |
| 0.2141 | 6.07 | 2000 | 0.3506 | 22.4675 |
| 0.2611 | 8.02 | 2500 | 0.3487 | 22.4124 |
Framework versions
- Transformers 4.37.0.dev0
- Pytorch 2.1.2+cu121
- Datasets 2.16.2.dev0
- Tokenizers 0.15.0