Downloads · 30 days
11
1% of all-time downloads
800min_whisper_small_FT_Algerian_Dialect
khier12/800min_whisper_small_FT_Algerian_Dialect
800min_whisper_small_FT_Algerian_Dialect is a automatic speech recognition model from khier12. 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
886
Public
Parameters
242M
83.4 GB on disk
Likes
1
Public
Repo makeup
Click a slice to open those files.
.safetensors967 MB · 81%
At a glance
- Task
- Automatic Speech Recognition
- Library
- transformers
- License
- apache-2.0
- Model type
- whisper
- Access
- Public
- Created
- Jan 24, 2026
- Updated
- Jun 5, 2026
- SHA
- 17e45482
Base models
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. -->
800min_whisper_small_FT_Algerian_Dialect
This model is a fine-tuned version of openai/whisper-small on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.7325
- Wer: 0.3873
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: 2
- eval_batch_size: 2
- seed: 42
- gradient_accumulation_steps: 16
- total_train_batch_size: 32
- optimizer: Use OptimizerNames.ADAMW_BNB 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: 400
- num_epochs: 12
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 47.9820 | 0.3655 | 100 | 1.6261 | 0.9067 |
| 25.3799 | 0.7311 | 200 | 1.1184 | 0.7250 |
| 13.3183 | 1.0950 | 300 | 0.7334 | 0.5913 |
| 10.1569 | 1.4606 | 400 | 0.6567 | 0.5504 |
| 9.7609 | 1.8261 | 500 | 0.6045 | 0.5910 |
| 7.0938 | 2.1901 | 600 | 0.5780 | 0.4806 |
| 6.2773 | 2.5556 | 700 | 0.5605 | 0.4492 |
| 6.1845 | 2.9212 | 800 | 0.5456 | 0.4197 |
| 4.0962 | 3.2851 | 900 | 0.5604 | 0.3930 |
| 4.0550 | 3.6507 | 1000 | 0.5475 | 0.4016 |
| 4.0792 | 4.0146 | 1100 | 0.5453 | 0.4018 |
| 2.6259 | 4.3802 | 1200 | 0.5656 | 0.3850 |
| 2.5990 | 4.7457 | 1300 | 0.5657 | 0.3807 |
| 2.5943 | 5.1097 | 1400 | 0.5835 | 0.3954 |
| 1.6042 | 5.4752 | 1500 | 0.5895 | 0.3897 |
| 1.7266 | 5.8408 | 1600 | 0.5925 | 0.3899 |
| 1.5761 | 6.2047 | 1700 | 0.6114 | 0.3970 |
| 0.9729 | 6.5703 | 1800 | 0.6162 | 0.3910 |
| 1.0095 | 6.9358 | 1900 | 0.6183 | 0.3951 |
| 0.7495 | 7.2997 | 2000 | 0.6421 | 0.3900 |
| 0.5854 | 7.6653 | 2100 | 0.6433 | 0.3860 |
| 0.6154 | 8.0292 | 2200 | 0.6487 | 0.3807 |
| 0.4020 | 8.3948 | 2300 | 0.6630 | 0.3912 |
| 0.3542 | 8.7603 | 2400 | 0.6704 | 0.3937 |
| 0.3408 | 9.1243 | 2500 | 0.6825 | 0.3895 |
| 0.2131 | 9.4898 | 2600 | 0.6918 | 0.3869 |
| 0.2272 | 9.8554 | 2700 | 0.6987 | 0.3902 |
| 0.1663 | 10.2193 | 2800 | 0.7089 | 0.3899 |
| 0.1338 | 10.5849 | 2900 | 0.7169 | 0.3886 |
| 0.1414 | 10.9504 | 3000 | 0.7181 | 0.3845 |
| 0.0960 | 11.3144 | 3100 | 0.7297 | 0.3871 |
| 0.0895 | 11.6799 | 3200 | 0.7325 | 0.3873 |
Framework versions
- Transformers 5.0.0
- Pytorch 2.11.0+cu128
- Datasets 4.0.0
- Tokenizers 0.22.2
- Task
- Automatic Speech Recognition
- Library
- transformers
- Type
- whisper
- License
- apache-2.0
- Created
- Jan 24, 2026
- Updated
- Jun 5, 2026