metadata
license: apache-2.0
tags:
- generated_from_trainer
datasets:
- fleurs
metrics:
- wer
model-index:
- name: whisper-large-v2-greek
results:
- task:
name: Automatic Speech Recognition
type: automatic-speech-recognition
dataset:
name: fleurs
type: fleurs
config: el_gr
split: test
args: el_gr
metrics:
- name: Wer
type: wer
value: 1.0564819086535293
whisper-large-v2-greek
This model is a fine-tuned version of openai/whisper-large-v2 on the fleurs dataset. It achieves the following results on the evaluation set:
- Loss: 0.2061
- Wer Ortho: 1.0424
- Wer: 1.0565
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: 2
- eval_batch_size: 2
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: constant_with_warmup
- lr_scheduler_warmup_steps: 50
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer Ortho | Wer |
|---|---|---|---|---|---|
| 0.1416 | 1.0 | 217 | 0.1611 | 1.2560 | 1.2638 |
| 0.0617 | 2.0 | 435 | 0.1612 | 1.1956 | 1.1930 |
| 0.027 | 3.0 | 653 | 0.1716 | 1.6495 | 1.6518 |
| 0.0155 | 4.0 | 871 | 0.1812 | 1.2816 | 1.2878 |
| 0.0114 | 5.0 | 1088 | 0.1792 | 1.0087 | 1.0071 |
| 0.0085 | 6.0 | 1306 | 0.1891 | 0.9757 | 0.9971 |
| 0.0073 | 7.0 | 1524 | 0.2017 | 1.0040 | 1.0225 |
| 0.0062 | 8.0 | 1742 | 0.1980 | 1.0737 | 1.0779 |
| 0.0094 | 9.0 | 1959 | 0.2103 | 0.8469 | 0.8459 |
| 0.0039 | 9.97 | 2170 | 0.2061 | 1.0424 | 1.0565 |
Framework versions
- Transformers 4.30.0.dev0
- Pytorch 2.0.1+cu117
- Datasets 2.13.1
- Tokenizers 0.13.3