Downloads · 30 days
2
10% of all-time downloads
ssc-mmc-mms-model-mix-adapt-max-lowlr
ctaguchi/ssc-mmc-mms-model-mix-adapt-max-lowlr
ssc-mmc-mms-model-mix-adapt-max-lowlr is a automatic speech recognition model from ctaguchi. 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
21
Public
Parameters
965M
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
- cc-by-nc-4.0
- Model type
- wav2vec2
- Access
- Public
- Created
- Dec 6, 2025
- Updated
- Dec 6, 2025
- SHA
- 986794ed
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. -->
ssc-mmc-mms-model-mix-adapt-max-lowlr
This model is a fine-tuned version of facebook/mms-1b-all on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.2748
- Cer: 0.2792
- Wer: 0.6328
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: 0.0003
- train_batch_size: 8
- eval_batch_size: 6
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 16
- 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: 100
- num_epochs: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Cer | Wer |
|---|---|---|---|---|---|
| 2.4444 | 0.3630 | 200 | 1.6953 | 0.3718 | 0.8006 |
| 1.6754 | 0.7260 | 400 | 1.4696 | 0.3340 | 0.7420 |
| 1.5773 | 1.0889 | 600 | 1.3822 | 0.3183 | 0.6998 |
| 1.4642 | 1.4519 | 800 | 1.3502 | 0.3136 | 0.7004 |
| 1.4418 | 1.8149 | 1000 | 1.3125 | 0.3030 | 0.6762 |
| 1.3594 | 2.1779 | 1200 | 1.3078 | 0.3016 | 0.6710 |
| 1.4188 | 2.5408 | 1400 | 1.3147 | 0.2995 | 0.6719 |
| 1.4157 | 2.9038 | 1600 | 1.2933 | 0.2984 | 0.6632 |
| 1.3717 | 3.2668 | 1800 | 1.2817 | 0.2905 | 0.6629 |
| 1.3127 | 3.6298 | 2000 | 1.2784 | 0.2905 | 0.6532 |
| 1.2914 | 3.9927 | 2200 | 1.2843 | 0.2907 | 0.6532 |
| 1.3149 | 4.3557 | 2400 | 1.2781 | 0.2884 | 0.6563 |
| 1.2927 | 4.7187 | 2600 | 1.2779 | 0.2867 | 0.6486 |
| 1.289 | 5.0817 | 2800 | 1.2766 | 0.2853 | 0.6441 |
| 1.2659 | 5.4446 | 3000 | 1.2808 | 0.2841 | 0.6454 |
| 1.2769 | 5.8076 | 3200 | 1.2634 | 0.2817 | 0.6385 |
| 1.2407 | 6.1706 | 3400 | 1.2764 | 0.2811 | 0.6447 |
| 1.2399 | 6.5336 | 3600 | 1.2775 | 0.2809 | 0.6392 |
| 1.2309 | 6.8966 | 3800 | 1.2719 | 0.2812 | 0.6385 |
| 1.2341 | 7.2595 | 4000 | 1.2717 | 0.2805 | 0.6375 |
| 1.2258 | 7.6225 | 4200 | 1.2705 | 0.2783 | 0.6346 |
| 1.2167 | 7.9855 | 4400 | 1.2715 | 0.2797 | 0.6338 |
| 1.2458 | 8.3485 | 4600 | 1.2731 | 0.2782 | 0.6344 |
| 1.1659 | 8.7114 | 4800 | 1.2721 | 0.2776 | 0.6317 |
| 1.1922 | 9.0744 | 5000 | 1.2733 | 0.2791 | 0.6336 |
| 1.1726 | 9.4374 | 5200 | 1.2761 | 0.2792 | 0.6359 |
| 1.2057 | 9.8004 | 5400 | 1.2748 | 0.2792 | 0.6328 |
Framework versions
- Transformers 4.57.2
- Pytorch 2.9.1+cu128
- Datasets 3.6.0
- Tokenizers 0.22.0
- Task
- Automatic Speech Recognition
- Library
- transformers
- Type
- wav2vec2
- License
- cc-by-nc-4.0
- Created
- Dec 6, 2025
- Updated
- Dec 6, 2025