Downloads · 30 days
12
3% of all-time downloads
mms-1b-algerian-east-v2
khier12/mms-1b-algerian-east-v2
mms-1b-algerian-east-v2 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 cc-by-nc-4.0.
should probably proofread and complete it, then remove this comment. --
All-time downloads
347
Public
Parameters
965M
81 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
- cc-by-nc-4.0
- Model type
- wav2vec2
- Access
- Public
- Created
- May 25, 2026
- Updated
- Jun 2, 2026
- SHA
- 69c5989b
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. -->
mms-1b-algerian-east-v2
This model is a fine-tuned version of khier12/800min_mms-1b_Full_FT_Algerian_Dialect on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.5946
- Wer: 0.4015
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: 1
- eval_batch_size: 1
- seed: 42
- gradient_accumulation_steps: 32
- 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: 150
- num_epochs: 12
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 134.4558 | 0.3656 | 100 | 1.6486 | 0.7299 |
| 102.5500 | 0.7312 | 200 | 1.2094 | 0.6529 |
| 33.0515 | 1.0951 | 300 | 1.0335 | 0.6089 |
| 33.1589 | 1.4606 | 400 | 0.9232 | 0.5719 |
| 31.2878 | 1.8262 | 500 | 0.8515 | 0.5467 |
| 29.0004 | 2.1901 | 600 | 0.8176 | 0.5312 |
| 27.0669 | 2.5557 | 700 | 0.7933 | 0.5148 |
| 26.3714 | 2.9213 | 800 | 0.7740 | 0.5059 |
| 27.1695 | 3.2852 | 900 | 0.7244 | 0.4826 |
| 27.6370 | 3.6507 | 1000 | 0.7020 | 0.4809 |
| 29.3353 | 4.0146 | 1100 | 0.7638 | 0.4757 |
| 26.7547 | 4.3802 | 1200 | 0.7195 | 0.4648 |
| 21.4594 | 4.7458 | 1300 | 0.6774 | 0.4547 |
| 29.6335 | 5.1097 | 1400 | 0.6554 | 0.4460 |
| 33.3210 | 5.4753 | 1500 | 0.6493 | 0.4443 |
| 25.4002 | 5.8409 | 1600 | 0.6405 | 0.4310 |
| 22.3004 | 6.2047 | 1700 | 0.6523 | 0.4296 |
| 28.1907 | 6.5703 | 1800 | 0.6510 | 0.4242 |
| 25.3794 | 6.9359 | 1900 | 0.6273 | 0.4216 |
| 23.8162 | 7.2998 | 2000 | 0.6147 | 0.4200 |
| 27.0182 | 7.6654 | 2100 | 0.6142 | 0.4156 |
| 20.3677 | 8.0292 | 2200 | 0.6578 | 0.4173 |
| 24.0046 | 8.3948 | 2300 | 0.6234 | 0.4104 |
| 18.2346 | 8.7604 | 2400 | 0.6196 | 0.4072 |
| 24.7345 | 9.1243 | 2500 | 0.6020 | 0.4085 |
| 20.5054 | 9.4899 | 2600 | 0.6006 | 0.4051 |
| 29.3066 | 9.8555 | 2700 | 0.5983 | 0.4029 |
| 16.9787 | 10.2194 | 2800 | 0.6045 | 0.4016 |
| 21.4099 | 10.5849 | 2900 | 0.5998 | 0.4029 |
| 15.6921 | 10.9505 | 3000 | 0.5976 | 0.4005 |
| 20.4868 | 11.3144 | 3100 | 0.5975 | 0.3997 |
| 27.2156 | 11.6800 | 3200 | 0.5946 | 0.4015 |
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
- wav2vec2
- License
- cc-by-nc-4.0
- Created
- May 25, 2026
- Updated
- Jun 2, 2026