metadata
library_name: transformers
language:
- pmx
license: apache-2.0
base_model: openai/whisper-small
tags:
- generated_from_trainer
datasets:
- iitd-duk/paula
metrics:
- wer
model-index:
- name: Whisper-Small-paula
results:
- task:
name: Automatic Speech Recognition
type: automatic-speech-recognition
dataset:
name: Paula
type: iitd-duk/paula
metrics:
- name: Wer
type: wer
value: 106.09137055837563
Whisper-Small-paula
This model is a fine-tuned version of openai/whisper-small on the Paula dataset. It achieves the following results on the evaluation set:
- Loss: 3.0264
- Wer: 106.0914
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: 12
- 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: 100
- training_steps: 1000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.3022 | 12.5 | 200 | 2.6624 | 109.5431 |
| 0.0077 | 25.0 | 400 | 2.8312 | 105.1777 |
| 0.0012 | 37.5 | 600 | 2.9571 | 110.4569 |
| 0.0008 | 50.0 | 800 | 3.0120 | 107.6142 |
| 0.0007 | 62.5 | 1000 | 3.0264 | 106.0914 |
Framework versions
- Transformers 4.46.0.dev0
- Pytorch 2.4.1+cu121
- Datasets 3.0.1
- Tokenizers 0.20.1