metadata
license: apache-2.0
tags:
- generated_from_trainer
datasets:
- common_voice_11_0
metrics:
- wer
model-index:
- name: emilios/whisper-medium-el-n2
results:
- task:
name: Automatic Speech Recognition
type: automatic-speech-recognition
dataset:
name: common_voice_11_0
type: common_voice_11_0
config: el
split: test
args: el
metrics:
- name: Wer
type: wer
value: 9.964710252600298
emilios/whisper-medium-el-n2
This model is a fine-tuned version of emilios/whisper-medium-el-n2 on the common_voice_11_0 dataset. It achieves the following results on the evaluation set:
- Loss: 0.5674
- Wer: 9.9647
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-06
- train_batch_size: 32
- eval_batch_size: 16
- seed: 42
- distributed_type: multi-GPU
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- training_steps: 10000
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.0014 | 58.82 | 1000 | 0.4951 | 10.3640 |
| 0.0006 | 117.65 | 2000 | 0.5181 | 10.2805 |
| 0.0007 | 175.82 | 3000 | 0.5317 | 10.1133 |
| 0.0004 | 234.65 | 4000 | 0.5396 | 10.1226 |
| 0.0004 | 293.47 | 5000 | 0.5532 | 10.1040 |
| 0.0013 | 352.29 | 6000 | 0.5645 | 10.0854 |
| 0.0002 | 411.12 | 7000 | 0.5669 | 10.1133 |
| 0.0001 | 469.94 | 8000 | 0.5669 | 9.8997 |
| 0.0001 | 528.76 | 9000 | 0.5645 | 9.9276 |
| 0.0001 | 587.82 | 10000 | 0.5674 | 9.9647 |
Framework versions
- Transformers 4.26.0.dev0
- Pytorch 2.0.0.dev20221216+cu116
- Datasets 2.7.1.dev0
- Tokenizers 0.13.2