Downloads · 30 days
2
10% of all-time downloads
mdd-arabic-e4-xlsr1b-specaugment-yaml-beam
omarZACK/mdd-arabic-e4-xlsr1b-specaugment-yaml-beam
mdd-arabic-e4-xlsr1b-specaugment-yaml-beam is a automatic speech recognition model from omarZACK. 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
20
Public
Parameters
963M
3.9 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors3.9 GB · 100%
At a glance
- Task
- Automatic Speech Recognition
- Library
- transformers
- License
- apache-2.0
- Model type
- wav2vec2
- Access
- Public
- Created
- Aug 21, 2026
- Updated
- Aug 22, 2026
- SHA
- f8da285d
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. -->
mdd-arabic-e4-xlsr1b-specaugment-yaml-beam
This model is a fine-tuned version of facebook/wav2vec2-xls-r-1b on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.0616
- Per: 0.2183
- Accuracy: 0.7657
- Precision: 0.2884
- Recall: 0.2785
- F1 Macro: 0.2831
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: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED 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: 2181
- num_epochs: 30
Training results
| Training Loss | Epoch | Step | Validation Loss | Per | Accuracy | Precision | Recall | F1 Macro |
|---|---|---|---|---|---|---|---|---|
| 0.9606 | 1.0 | 728 | 0.8225 | 0.1938 | 0.7931 | 0.3083 | 0.3052 | 0.3054 |
| 0.4990 | 2.0 | 1456 | 0.8834 | 0.1838 | 0.7953 | 0.3109 | 0.2929 | 0.3013 |
| 0.3630 | 3.0 | 2184 | 1.0047 | 0.1990 | 0.7825 | 0.3040 | 0.2791 | 0.2905 |
| 0.4224 | 4.0 | 2912 | 0.9818 | 0.2031 | 0.7810 | 0.2992 | 0.2923 | 0.2953 |
| 0.3411 | 5.0 | 3640 | 1.1478 | 0.2135 | 0.7705 | 0.2897 | 0.2764 | 0.2828 |
| 0.3553 | 6.0 | 4368 | 1.0590 | 0.2170 | 0.7693 | 0.2923 | 0.2849 | 0.2882 |
| 0.3267 | 7.0 | 5096 | 1.0616 | 0.2183 | 0.7657 | 0.2884 | 0.2785 | 0.2831 |
Framework versions
- Transformers 5.15.0
- Pytorch 2.13.0+cu130
- Datasets 5.0.1
- Tokenizers 0.22.2
- Task
- Automatic Speech Recognition
- Library
- transformers
- Type
- wav2vec2
- License
- apache-2.0
- Created
- Aug 21, 2026
- Updated
- Aug 22, 2026