Downloads · 30 days
7
26% of all-time downloads
Kabardian-ASR-colab3
MatricariaV/Kabardian-ASR-colab3
Kabardian-ASR-colab3 is a automatic speech recognition model from MatricariaV. 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
27
Public
Parameters
965M
123 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
- Feb 24, 2025
- Updated
- Feb 24, 2025
- SHA
- 8dc24c2d
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. -->
Kabardian-ASR-colab3
This model is a fine-tuned version of facebook/mms-1b-all on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.1713
- Wer: 0.3087
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: 0.001
- train_batch_size: 8
- eval_batch_size: 8
- 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
- lr_scheduler_warmup_steps: 100
- num_epochs: 6
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.8505 | 0.1871 | 200 | 0.4842 | 0.8062 |
| 0.6677 | 0.3742 | 400 | 0.4299 | 0.7603 |
| 0.5938 | 0.5613 | 600 | 0.3595 | 0.6693 |
| 0.5221 | 0.7484 | 800 | 0.3258 | 0.6032 |
| 0.5563 | 0.9355 | 1000 | 0.3061 | 0.5789 |
| 0.5082 | 1.1225 | 1200 | 0.2644 | 0.4787 |
| 0.5271 | 1.3096 | 1400 | 0.2666 | 0.4908 |
| 0.4715 | 1.4967 | 1600 | 0.2637 | 0.4558 |
| 0.5146 | 1.6838 | 1800 | 0.2508 | 0.4528 |
| 0.4727 | 1.8709 | 2000 | 0.2383 | 0.4220 |
| 0.467 | 2.0580 | 2200 | 0.2298 | 0.4124 |
| 0.449 | 2.2451 | 2400 | 0.2245 | 0.4014 |
| 0.4389 | 2.4322 | 2600 | 0.2138 | 0.3778 |
| 0.4332 | 2.6193 | 2800 | 0.2122 | 0.3919 |
| 0.4239 | 2.8064 | 3000 | 0.2126 | 0.4074 |
| 0.4228 | 2.9935 | 3200 | 0.2019 | 0.3658 |
| 0.4119 | 3.1805 | 3400 | 0.2086 | 0.3789 |
| 0.4034 | 3.3676 | 3600 | 0.1985 | 0.3632 |
| 0.4098 | 3.5547 | 3800 | 0.2000 | 0.3536 |
| 0.3788 | 3.7418 | 4000 | 0.1974 | 0.3686 |
| 0.3758 | 3.9289 | 4200 | 0.2005 | 0.3947 |
| 0.4135 | 4.1160 | 4400 | 0.1840 | 0.3492 |
| 0.3503 | 4.3031 | 4600 | 0.1830 | 0.3373 |
| 0.3556 | 4.4902 | 4800 | 0.1788 | 0.3253 |
| 0.3227 | 4.6773 | 5000 | 0.1804 | 0.3234 |
| 0.3259 | 4.8644 | 5200 | 0.1778 | 0.3228 |
| 0.3577 | 5.0514 | 5400 | 0.1789 | 0.3248 |
| 0.3307 | 5.2385 | 5600 | 0.1752 | 0.3161 |
| 0.346 | 5.4256 | 5800 | 0.1756 | 0.3114 |
| 0.329 | 5.6127 | 6000 | 0.1728 | 0.3162 |
| 0.3188 | 5.7998 | 6200 | 0.1719 | 0.3167 |
| 0.3149 | 5.9869 | 6400 | 0.1713 | 0.3087 |
Framework versions
- Transformers 4.50.0.dev0
- Pytorch 2.5.1+cu121
- Datasets 3.3.2
- Tokenizers 0.21.0
- Task
- Automatic Speech Recognition
- Library
- transformers
- Type
- wav2vec2
- License
- cc-by-nc-4.0
- Created
- Feb 24, 2025
- Updated
- Feb 24, 2025