Downloads · 30 days
6
20% of all-time downloads
whisper-small-benin
Sammyeazzy/whisper-small-benin
whisper-small-benin is a automatic speech recognition model from Sammyeazzy. Use it when you need speech turned into text. It is set up for transformers. The card lists the license as apache-2.0.
should probably proofread and complete it, then remove this comment. --
All-time downloads
30
Public
Parameters
242M
5.8 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors967 MB · 100%
At a glance
- Task
- Automatic Speech Recognition
- Library
- transformers
- License
- apache-2.0
- Model type
- whisper
- Access
- Public
- Created
- Sep 2, 2025
- Updated
- Sep 8, 2025
- SHA
- ebb755bd
Base models
Datasets
Model card
From the Hugging Face model README
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
Whisper Small Benin- Samuel Easy
This model is a fine-tuned version of openai/whisper-small on the Benin_NLP dataset. It achieves the following results on the evaluation set:
- Loss: 0.3766
- Wer: 88.8889
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: 0.0001
- train_batch_size: 2
- eval_batch_size: 2
- 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
- num_epochs: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.8211 | 0.2632 | 5 | 1.8901 | 4083.3333 |
| 3.1022 | 0.5263 | 10 | 3.5744 | 88.8889 |
| 1.4135 | 0.7895 | 15 | 1.0322 | 111.1111 |
| 0.4287 | 1.0526 | 20 | 1.4105 | 88.8889 |
| 1.0782 | 1.3158 | 25 | 1.1904 | 100.0 |
| 1.7946 | 1.5789 | 30 | 1.0665 | 111.1111 |
| 0.9281 | 1.8421 | 35 | 0.6988 | 88.8889 |
| 0.5404 | 2.1053 | 40 | 0.4681 | 105.5556 |
| 0.8641 | 2.3684 | 45 | 0.4852 | 111.1111 |
| 0.4218 | 2.6316 | 50 | 0.5601 | 100.0 |
| 0.5228 | 2.8947 | 55 | 0.4776 | 100.0 |
| 0.6757 | 3.1579 | 60 | 0.3882 | 94.4444 |
| 0.7486 | 3.4211 | 65 | 0.5367 | 105.5556 |
| 0.4112 | 3.6842 | 70 | 0.3424 | 88.8889 |
| 0.3417 | 3.9474 | 75 | 0.4473 | 100.0 |
| 0.6627 | 4.2105 | 80 | 0.5523 | 111.1111 |
| 0.4566 | 4.4737 | 85 | 0.5891 | 88.8889 |
| 0.934 | 4.7368 | 90 | 0.4701 | 100.0 |
| 0.4693 | 5.0 | 95 | 0.4066 | 100.0 |
| 0.4143 | 5.2632 | 100 | 0.3859 | 100.0 |
| 0.6106 | 5.5263 | 105 | 0.4182 | 88.8889 |
| 0.6801 | 5.7895 | 110 | 0.4446 | 100.0 |
| 0.371 | 6.0526 | 115 | 0.4075 | 88.8889 |
| 0.4851 | 6.3158 | 120 | 0.4135 | 88.8889 |
| 0.4554 | 6.5789 | 125 | 0.3927 | 111.1111 |
| 0.3334 | 6.8421 | 130 | 0.3963 | 100.0 |
| 0.385 | 7.1053 | 135 | 0.3853 | 100.0 |
| 0.5169 | 7.3684 | 140 | 0.3630 | 100.0 |
| 0.4169 | 7.6316 | 145 | 0.3652 | 83.3333 |
| 0.4318 | 7.8947 | 150 | 0.3792 | 100.0 |
| 0.3229 | 8.1579 | 155 | 0.4093 | 100.0 |
| 0.3431 | 8.4211 | 160 | 0.4110 | 100.0 |
| 0.4228 | 8.6842 | 165 | 0.3982 | 100.0 |
| 0.3957 | 8.9474 | 170 | 0.3860 | 88.8889 |
| 0.5902 | 9.2105 | 175 | 0.3858 | 88.8889 |
| 0.468 | 9.4737 | 180 | 0.3835 | 88.8889 |
| 0.397 | 9.7368 | 185 | 0.3810 | 88.8889 |
| 0.4073 | 10.0 | 190 | 0.3766 | 88.8889 |
Framework versions
- Transformers 4.52.0
- Pytorch 2.8.0+cu126
- Datasets 3.6.0
- Tokenizers 0.21.4
- Task
- Automatic Speech Recognition
- Library
- transformers
- Type
- whisper
- License
- apache-2.0
- Languages
- ben
- Created
- Sep 2, 2025
- Updated
- Sep 8, 2025