metadata
license: apache-2.0
base_model: openai/whisper-small
tags:
- generated_from_trainer
datasets:
- common_voice_17_0
metrics:
- wer
model-index:
- name: whisper-small-malayalam-colab-CV17.0
results:
- task:
name: Automatic Speech Recognition
type: automatic-speech-recognition
dataset:
name: common_voice_17_0
type: common_voice_17_0
config: ml
split: test
args: ml
metrics:
- name: Wer
type: wer
value: 0.6534493874919407
whisper-small-malayalam-colab-CV17.0
This model is a fine-tuned version of openai/whisper-small on the common_voice_17_0 dataset. It achieves the following results on the evaluation set:
- Loss: 0.3197
- Wer: 0.6534
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-05
- train_batch_size: 16
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.15
- training_steps: 2000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.7367 | 1.5748 | 200 | 0.2861 | 0.8443 |
| 0.1405 | 3.1496 | 400 | 0.2516 | 0.7550 |
| 0.061 | 4.7244 | 600 | 0.2315 | 0.7121 |
| 0.0295 | 6.2992 | 800 | 0.2600 | 0.6995 |
| 0.0161 | 7.8740 | 1000 | 0.2731 | 0.6721 |
| 0.0073 | 9.4488 | 1200 | 0.2925 | 0.6847 |
| 0.0033 | 11.0236 | 1400 | 0.3144 | 0.6692 |
| 0.0014 | 12.5984 | 1600 | 0.3111 | 0.6580 |
| 0.0002 | 14.1732 | 1800 | 0.3161 | 0.6557 |
| 0.0001 | 15.7480 | 2000 | 0.3197 | 0.6534 |
Framework versions
- Transformers 4.42.4
- Pytorch 2.4.0+cu121
- Datasets 2.21.0
- Tokenizers 0.19.1