Downloads · 30 days
1
5% of all-time downloads
xls-r-300m-gaphon-lowbatch
duck-hug-567/xls-r-300m-gaphon-lowbatch
xls-r-300m-gaphon-lowbatch is a automatic speech recognition model from duck-hug-567. 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
316M
37.9 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors1.3 GB · 100%
At a glance
- Task
- Automatic Speech Recognition
- Library
- transformers
- License
- apache-2.0
- Model type
- wav2vec2
- Access
- Public
- Created
- Apr 15, 2025
- Updated
- Apr 15, 2025
- SHA
- 5fe954b0
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. -->
xls-r-300m-gaphon-lowbatch
This model is a fine-tuned version of facebook/wav2vec2-xls-r-300m on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.5648
- Cer: 0.1371
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: 3e-05
- train_batch_size: 4
- eval_batch_size: 8
- seed: 42
- optimizer: Use 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: 2000
- num_epochs: 30
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Cer |
|---|---|---|---|---|
| 2.676 | 1.0 | 3827 | 2.0633 | 0.5072 |
| 1.8586 | 2.0 | 7654 | 0.9506 | 0.2137 |
| 1.7579 | 3.0 | 11481 | 0.7871 | 0.1926 |
| 1.6515 | 4.0 | 15308 | 0.7346 | 0.1836 |
| 1.5873 | 5.0 | 19135 | 0.6987 | 0.1770 |
| 1.5397 | 6.0 | 22962 | 0.6661 | 0.1610 |
| 1.4802 | 7.0 | 26789 | 0.6506 | 0.1655 |
| 1.4505 | 8.0 | 30616 | 0.6301 | 0.1625 |
| 1.3571 | 9.0 | 34443 | 0.6343 | 0.1646 |
| 1.4287 | 10.0 | 38270 | 0.6134 | 0.1540 |
| 1.3994 | 11.0 | 42097 | 0.6053 | 0.1513 |
| 1.3915 | 12.0 | 45924 | 0.5928 | 0.1513 |
| 1.4206 | 13.0 | 49751 | 0.6046 | 0.1459 |
| 1.3111 | 14.0 | 53578 | 0.5848 | 0.1489 |
| 1.2982 | 15.0 | 57405 | 0.5825 | 0.1470 |
| 1.2392 | 16.0 | 61232 | 0.5775 | 0.1472 |
| 1.1944 | 17.0 | 65059 | 0.5772 | 0.1425 |
| 1.2184 | 18.0 | 68886 | 0.5686 | 0.1425 |
| 1.2005 | 19.0 | 72713 | 0.5651 | 0.1437 |
| 1.1775 | 20.0 | 76540 | 0.5717 | 0.1404 |
| 1.2213 | 21.0 | 80367 | 0.5688 | 0.1412 |
| 1.1954 | 22.0 | 84194 | 0.5600 | 0.1417 |
| 1.229 | 23.0 | 88021 | 0.5598 | 0.1409 |
| 1.1695 | 24.0 | 91848 | 0.5715 | 0.1399 |
| 1.1637 | 25.0 | 95675 | 0.5656 | 0.1406 |
| 1.0601 | 26.0 | 99502 | 0.5683 | 0.1398 |
| 1.084 | 27.0 | 103329 | 0.5681 | 0.1394 |
| 1.1967 | 28.0 | 107156 | 0.5683 | 0.1362 |
| 1.1312 | 29.0 | 110983 | 0.5668 | 0.1381 |
| 1.1139 | 30.0 | 114810 | 0.5648 | 0.1371 |
Framework versions
- Transformers 4.47.0
- Pytorch 2.5.1+cu121
- Datasets 3.3.1
- Tokenizers 0.21.0
- Task
- Automatic Speech Recognition
- Library
- transformers
- Type
- wav2vec2
- License
- apache-2.0
- Created
- Apr 15, 2025
- Updated
- Apr 15, 2025