metadata
language:
- ru
license: apache-2.0
tags:
- generated_from_trainer
base_model: openai/whisper-tiny
datasets:
- bond005/podlodka_speech
metrics:
- wer
model-index:
- name: whisper-tiny-ru
results:
- task:
type: automatic-speech-recognition
name: Automatic Speech Recognition
dataset:
name: Podlodka Speech
type: bond005/podlodka_speech
args: 'config: ru, split: test'
metrics:
- type: wer
value: 99.38757655293088
name: Wer
whisper-tiny-ru
This model is a fine-tuned version of openai/whisper-tiny on the Podlodka Speech dataset. It achieves the following results on the evaluation set:
- Loss: 1.4991
- Wer: 99.3876
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: 8
- eval_batch_size: 4
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 100
- training_steps: 500
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.295 | 5.5556 | 50 | 1.1982 | 83.2896 |
| 0.1297 | 11.1111 | 100 | 1.2768 | 76.0280 |
| 0.0517 | 16.6667 | 150 | 1.3594 | 72.5284 |
| 0.0203 | 22.2222 | 200 | 1.3969 | 85.4768 |
| 0.0094 | 27.7778 | 250 | 1.4394 | 104.2870 |
| 0.0061 | 33.3333 | 300 | 1.4646 | 87.8390 |
| 0.0049 | 38.8889 | 350 | 1.4813 | 90.4637 |
| 0.0043 | 44.4444 | 400 | 1.4909 | 86.7017 |
| 0.004 | 50.0 | 450 | 1.4973 | 99.6500 |
| 0.0038 | 55.5556 | 500 | 1.4991 | 99.3876 |
Framework versions
- Transformers 4.40.2
- Pytorch 2.2.1+cu121
- Datasets 2.19.1
- Tokenizers 0.19.1