metadata
library_name: transformers
language:
- id
license: apache-2.0
base_model: kneth90/whisper-small-id-kp
tags:
- generated_from_trainer
datasets:
- kneth90/snamol_temp
metrics:
- wer
model-index:
- name: WHISPER Small KP
results:
- task:
name: Automatic Speech Recognition
type: automatic-speech-recognition
dataset:
name: Common Voice 11.0
type: kneth90/snamol_temp
args: 'config: hi, split: test'
metrics:
- name: Wer
type: wer
value: 814.2857142857142
WHISPER Small KP
This model is a fine-tuned version of kneth90/whisper-small-id-kp on the Common Voice 11.0 dataset. It achieves the following results on the evaluation set:
- Loss: 1.4420
- Wer: 814.2857
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 OptimizerNames.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: 6000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.0 | 1000.0 | 1000 | 0.5975 | 7.1429 |
| 0.0 | 2000.0 | 2000 | 0.8193 | 14.2857 |
| 0.0 | 3000.0 | 3000 | 1.0415 | 50.0 |
| 0.0 | 4000.0 | 4000 | 1.3226 | 50.0 |
| 0.0 | 5000.0 | 5000 | 1.3835 | 42.8571 |
| 0.0 | 6000.0 | 6000 | 1.4420 | 814.2857 |
Framework versions
- Transformers 4.48.1
- Pytorch 2.5.1+cu124
- Datasets 3.2.0
- Tokenizers 0.21.0