metadata
library_name: transformers
language:
- ne
license: apache-2.0
base_model: openai/whisper-large-v3
tags:
- generated_from_trainer
datasets:
- kiranpantha/OpenSLR54-Balanced-Nepali
metrics:
- wer
model-index:
- name: Whisper Large v3 Nepali - Kiran Pantha
results:
- task:
name: Automatic Speech Recognition
type: automatic-speech-recognition
dataset:
name: OpenSLR54
type: kiranpantha/OpenSLR54-Balanced-Nepali
config: default
split: test
args: 'config: ne, split: test'
metrics:
- name: Wer
type: wer
value: 20.48611111111111
Whisper Large v3 Nepali - Kiran Pantha
This model is a fine-tuned version of openai/whisper-large-v3 on the OpenSLR54 dataset. It achieves the following results on the evaluation set:
- Loss: 0.1507
- Wer: 20.4861
- Cer: 4.9839
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: 16
- eval_batch_size: 8
- seed: 42
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- training_steps: 5000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Cer | Validation Loss | Wer |
|---|---|---|---|---|---|
| 0.209 | 0.3597 | 300 | 10.9916 | 0.2021 | 41.1574 |
| 0.1714 | 0.7194 | 600 | 8.9474 | 0.1755 | 35.7176 |
| 0.101 | 1.0791 | 900 | 7.5565 | 0.1485 | 29.6991 |
| 0.0902 | 1.4388 | 1200 | 7.2372 | 0.1396 | 28.2407 |
| 0.0872 | 1.7986 | 1500 | 7.8024 | 0.1319 | 27.9861 |
| 0.0453 | 2.1583 | 1800 | 6.3344 | 0.1374 | 26.2269 |
| 0.0368 | 2.5180 | 2100 | 6.1766 | 0.1381 | 25.2315 |
| 0.0472 | 2.8777 | 2400 | 5.8316 | 0.1316 | 24.1435 |
| 0.0191 | 3.2374 | 2700 | 5.8059 | 0.1356 | 24.0278 |
| 0.0185 | 3.5971 | 3000 | 5.5674 | 0.1376 | 23.125 |
| 0.0182 | 3.9568 | 3300 | 5.5123 | 0.1360 | 23.0556 |
| 0.0074 | 4.3165 | 3600 | 5.2077 | 0.1428 | 21.7130 |
| 0.0086 | 4.6763 | 3900 | 5.1784 | 0.1433 | 21.2731 |
| 0.0031 | 5.0360 | 4200 | 0.1421 | 21.1806 | 5.0279 |
| 0.0024 | 5.3957 | 4500 | 0.1482 | 20.7870 | 4.9912 |
| 0.0014 | 5.7554 | 4800 | 0.1507 | 20.4861 | 4.9839 |
Framework versions
- Transformers 4.46.3
- Pytorch 2.5.1+cxx11.abi
- Datasets 3.2.0
- Tokenizers 0.20.3