metadata
library_name: transformers
language:
- wo
license: apache-2.0
base_model: openai/whisper-small
tags:
- generated_from_trainer
datasets:
- IndabaxSenegal/asr-wolof-dataset
metrics:
- wer
model-index:
- name: Whisper small Wolof
results:
- task:
name: Automatic Speech Recognition
type: automatic-speech-recognition
dataset:
name: ASR Wolof Dataset
type: IndabaxSenegal/asr-wolof-dataset
args: 'config: wo, split: test'
metrics:
- name: Wer
type: wer
value: 51.15567641060503
Whisper small Wolof
This model is a fine-tuned version of openai/whisper-small on the ASR Wolof Dataset dataset. It achieves the following results on the evaluation set:
- Loss: 0.8271
- Wer: 51.1557
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: 8
- eval_batch_size: 4
- 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 | Validation Loss | Wer |
|---|---|---|---|---|
| 0.1739 | 0.1724 | 1000 | 0.8788 | 51.3684 |
| 0.1841 | 0.3448 | 2000 | 0.8774 | 52.4196 |
| 0.199 | 0.5172 | 3000 | 0.8464 | 54.0335 |
| 0.2244 | 0.6897 | 4000 | 0.8384 | 49.9673 |
| 0.1751 | 0.8621 | 5000 | 0.8271 | 51.1557 |
Framework versions
- Transformers 4.46.3
- Pytorch 2.4.0
- Datasets 3.1.0
- Tokenizers 0.20.0